Information processing device and program

Abstract

To provide an information processing apparatus and a program capable of early predicting products whose demand is expected to increase in the future.SOLUTION: A demand prediction apparatus (an information processing apparatus) comprises: a sale increase product detection unit (detection unit) for detecting a product whose number of sales tends to increase in a wide area; a search number acquisition unit (acquisition unit) for acquiring the number of searches of the product from a preset first information source; a review type analysis unit (calculation unit) for calculating a proportion of positive reviews about the product from reviews posted to the first information source; a demand prediction unit for determining whether wide demand for the product can be expected on the basis of the number of searches acquired by the search number acquisition unit and the proportion of the positive reviews calculated by the review type analysis unit; and a prediction result output unit (output unit) for outputting the determination result of the demand prediction unit.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present invention relates to an information processing device and a program. [Background technology]

[0002] Conventionally, when a product with higher sales than usual is detected, an information processing device is known that performs an information search using the product name as a keyword, retrieves pages of television programs, information sites, news sites, etc. that are presumed to have contributed to the increase in sales, and displays these pages on the display of a POS terminal (for example, Patent Document 1).

[0003] It is desirable for product ordering personnel to be able to quickly identify products that have been featured on television or the like and are experiencing a sudden increase in demand, or products for which demand is increasing locally only in areas near a store, predict the likelihood of an increase in demand, and place appropriate orders. Applying the technology disclosed in Patent Document 1 makes it possible to estimate the information source associated with an increase in sales volume, but it does not identify the reason for the increase in sales volume, so there is insufficient information for ordering personnel to make a decision to increase the order quantity. Summary of the Invention [Problem to be solved by the invention]

[0004] The problem to be solved by the present invention is to provide an information processing device and a program that can quickly predict products for which demand is expected to increase in the future. [Means for solving the problem]

[0005] An information processing apparatus according to an embodiment includes a detection unit, an acquisition unit, a calculation unit, a demand forecasting unit, and an output unit. Comparing the current sales volume of a product with the past sales volume, based on whether the sales volume growth rate exceeds a predetermined threshold, The acquisition unit detects products that are showing a tendency to increase in sales. And Ta It is expected to be widely accessed from all over the country.The number of searches for the product is acquired from the first information source. The calculation unit calculates the proportion of positive opinions about the product from the opinions submitted to the first information source. The demand forecasting unit calculates the number of searches acquired by the acquisition unit. Increase in and the percentage of positive opinions calculated by the calculation department. , whether the sum of the integrated values ​​for each first information source is equal to or greater than a threshold value. The output unit outputs the determination result of the demand forecasting unit. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a block diagram showing an example of a schematic configuration of a demand forecasting system. [Figure 2] FIG. 2 is a hardware block diagram showing an example of the hardware configuration of a demand prediction device included in the demand prediction system. [Figure 3] FIG. 3 is a diagram illustrating an example of the data structure of the product sales volume data. [Figure 4] FIG. 4 is a diagram illustrating an example of the data structure of the specific user purchase data. [Figure 5] FIG. 5 is a diagram illustrating an example of the data structure of the product search count data. [Figure 6] FIG. 6 is a diagram illustrating an example of a data structure of opinion type data. [Figure 7] FIG. 7 is a diagram illustrating an example of a data structure of the demand forecast data. [Figure 8] FIG. 8 is a functional block diagram illustrating an example of a functional configuration of the demand prediction device. [Figure 9] FIG. 9 is a diagram illustrating a method by which the demand prediction device predicts whether a specific product is likely to be in widespread demand in the future. [Figure 10] FIG. 10 is a diagram illustrating a method by which the demand forecasting device predicts whether demand for a specific product is expected in a specific region in the future. [Figure 11] FIG. 11 is a diagram showing an example of a product ordering screen in which the prediction results of the demand prediction device are output. [Figure 12] FIG. 12 is a diagram illustrating an example of a wide-area demand forecast result output by the demand forecasting device. [Figure 13] FIG. 13 is a diagram illustrating an example of a local demand forecast result output by the demand forecasting device. [Figure 14] FIG. 14 is a flowchart showing an example of the flow of processing performed by the demand prediction device. [Figure 15] FIG. 15 is a flowchart showing an example of the flow of the wide-area demand forecasting process. [Figure 16] FIG. 16 is a flowchart showing an example of the flow of the local demand forecasting process. DETAILED DESCRIPTION OF THE INVENTION

[0007] An embodiment in which an information processing device of the present invention is applied to a demand forecasting system 10 will be described with reference to the drawings.

[0008] (Outline of the demand forecasting system) A schematic configuration of a demand forecasting system 10 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of a schematic configuration of a demand forecasting system.

[0009] The demand forecasting system 10 predicts products for which demand is likely to increase over a wide area. The demand forecasting system 10 also predicts products for which demand is likely to increase in a limited area (local area) such as a specific region. The demand forecasting system 10 includes a demand forecasting device 12, a first information source 14 and a second information source 15 that exist on a network N such as the Internet connected to the demand forecasting device 12, a sales management server 16 connected to the demand forecasting device 12 via the network N, and a POS terminal 18 and a POS terminal 19 that are connected to the sales management server 16 via a network such as a LAN.

[0010] The demand forecasting device 12 predicts products whose demand is likely to increase over a wide area based on information obtained from a first information source 14 connected to the network N and product sales information obtained from a sales management server 16 (hereinafter referred to as wide-area demand forecasting). The demand forecasting device 12 also predicts products whose demand is likely to increase in a specific area based on information obtained from a second information source 15 connected to the network N and product sales information obtained from a POS terminal 18 (hereinafter referred to as local demand forecasting). The demand forecasting device 12 is an example of an information processing device in the present disclosure.

[0011] An operator of the demand forecasting system 10 places an order for a product by referring to the forecast results of the demand forecasting device 12. Therefore, in the following description, the demand forecasting device 12 will be described as including a product ordering function.

[0012] The first information source 14 is a homepage, search service, or SNS (Social Networking Service) site such as an image posting site, a video posting site, or a message posting site that can be accessed via the network N. In particular, the first information source 14 is an information source that is not limited to a particular region and is expected to be accessed widely from all over the country.

[0013] The second information source 15 is a homepage of a city hall A, a school B, or the like that is specialized for a specific region and can be accessed via the network N. In other words, the second information source 15 is an information source that is expected to be accessed frequently from a specific region.

[0014] The POS terminal 18 is a terminal device that is installed in a store in a predetermined specific area and that manages the sale of goods.

[0015] The POS terminal 19 is a terminal device that is installed in an area other than a specific area and that manages the sale of goods.

[0016] Sales management server 16 is a server device that comprehensively manages POS terminals 18 and 19. Sales management server 16 acquires and manages sales information of products at POS terminals 18 and 19. The number of sales management servers 16 is not limited to one, and there may be a configuration in which there are multiple sales management servers 16 that each manage the POS terminals in each region, for example.

[0017] (Hardware configuration of demand forecasting device) The hardware configuration of the demand prediction device 12 will be described with reference to Fig. 2. Fig. 2 is a hardware block diagram showing an example of the hardware configuration of the demand prediction device included in the demand prediction system.

[0018] The demand forecasting device 12 includes a control unit 21 for controlling each unit of the demand forecasting device 12. The control unit 21 includes a CPU (Central Processing Unit) 22, a ROM (Read Only Memory) 23, and a RAM (Random Access Memory) 24. The CPU 22 is connected to the ROM 23 and the RAM 24 via an internal bus 33 such as an address bus and a data bus. The CPU 22 loads various programs stored in the ROM 23 and the storage unit 25 into the RAM 24. The CPU 22 controls the operation of the demand forecasting device 12 by operating in accordance with the various programs loaded into the RAM 24. In other words, the control unit 21 has the configuration of a typical computer.

[0019] The control unit 21 is connected to the storage unit 25, the display device 34, the operation device 35, and the communication interface 36 via an internal bus 33.

[0020] The storage unit 25 is a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). Alternatively, the storage unit 25 may be a non-volatile memory such as a flash memory that retains stored information even when the power is turned off. The storage unit 25 stores a control program 26 required for the demand prediction device 12 to perform predetermined operations, and various data required for the demand prediction device 12 to perform various demand predictions, such as product sales volume data 27, specific user purchase data 28, product search volume data 29, opinion type data 30, demand prediction data 31, and syntax analysis knowledge data 32.

[0021] The control program 26 is a program that controls the overall operation of the demand forecasting device 12 .

[0022] The product sales volume data 27 is data that records, for example, the number of products sold nationwide by product code and date, which is tallied by the sales management server 16. The data structure of the product sales volume data 27 will be described in detail later (see FIG. 3).

[0023] The specific user purchase data 28 is, for example, product purchase data of users residing in a specific region. In this embodiment, the specific user purchase data 28 is purchase data of users who use an application that can electronically register products and make payments when purchasing products at a store using their own mobile terminal, and whose address declared when registering to use the application falls within the specific region. The data structure of the specific user purchase data 28 will be described in detail later (see FIG. 4). The scope of the specific user purchase data 28 is not limited to this, and may be, for example, product purchase data of all users at a specific store within a specific region.

[0024] Product search count data 29 is data that records the number of searches for products whose sales are on the rise for each date. The information sources to be searched are the first information source 14 and the second information source 15. The data structure of product search count data 29 will be described in detail later (see FIG. 5).

[0025] The opinion type data 30 is data that identifies the content of opinions posted to the first information source 14 and the second information source 15 regarding products whose sales are on the rise. More specifically, the opinion type data 30 indicates the results of classifying opinions posted regarding products whose sales are on the rise as positive opinions, negative opinions, or neutral opinions. The data structure of the opinion type data 30 will be described in detail later (see FIG. 6).

[0026] The demand forecast data 31 is data that records the results of an evaluation of whether demand for a product whose sales volume is on the rise can be expected widely or whether demand for the product can be expected in a specific region. The data structure of the demand forecast data 31 will be described in detail later (see FIG. 7).

[0027] The syntactic analysis knowledge data 32 is data used to identify the content of opinions posted on the first information source 14 and the second information source 15. In order to analyze the structure of a sentence that expresses an opinion, the syntactic analysis knowledge data 32 includes rules for dividing a sentence into words, rules for analyzing the dependencies between the divided words, and rules for identifying whether the opinion is positive or negative. Note that syntactic analysis is a technique that is widely used in analyzing the content of comments posted on web pages, and therefore a detailed explanation will be omitted.

[0028] The display device 34 displays a screen generated in response to an instruction from the control unit 21. The display device 34 is configured by a device such as an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display.

[0029] The operation device 35 acquires an input operation by the operator and transmits it to the control unit 21. The operation device 35 is configured with devices such as a touch panel and a keyboard, for example.

[0030] The communication interface 36 is an interface for accessing the first information source 14 and the second information source 15 via the network N. The communication interface 36 is also an interface for accessing the POS terminal 18, the POS terminal 19, and the sales management server 16.

[0031] (Data structure of product sales data) The data structure of the product sales volume data 27 will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the data structure of the product sales volume data.

[0032] As shown in FIG. 3, the product sales volume data 27 stores the date and the sales volume for each product code that uniquely identifies the product, in association with each other.

[0033] The demand forecasting device 12 acquires sales volume information from the sales management server 16.

[0034] (Data structure of specific user purchase data) The data structure of the specific user purchase data 28 will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the data structure of the specific user purchase data.

[0035] As shown in FIG. 4, the specific user purchase data 28 stores purchase information of each customer for each customer code that uniquely identifies a customer whose address is in a specific area.

[0036] The purchase information includes the date the product was purchased, the store code where the product was purchased, the product code where the product was purchased, the list price of the purchased product, and the purchase amount.

[0037] (Data structure of product search count data) The data structure of the product search count data 29 will be described with reference to Fig. 5. Fig. 5 is a diagram showing an example of the data structure of the product search count data.

[0038] As shown in FIG. 5, the product search count data 29 stores, for each product code that uniquely identifies a product, search count information for the product having that product code.

[0039] The search count information includes an information source ID that uniquely identifies the information source of the search, the date the search was performed, and the number of searches. The information source ID is identification information that uniquely identifies the first information source 14 and the second information source 15.

[0040] (Data structure of opinion type data) The data structure of the opinion type data 30 will be described with reference to Fig. 6. Fig. 6 is a diagram showing an example of the data structure of the opinion type data.

[0041] As shown in FIG. 6, the opinion type data 30 stores, for each product code that uniquely identifies a product, opinion identification information that identifies an opinion related to the product posted to the information source.

[0042] The opinion identification information includes the date the opinion was posted, an information source ID that uniquely identifies the information source to which the opinion was posted, the number of positive opinions, the number of negative opinions, and the number of opinions that are neither positive nor negative.

[0043] (Data structure of demand forecast data) The data structure of the demand forecast data 31 will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of the data structure of the demand forecast data.

[0044] As shown in FIG. 7, the demand forecast data 31 stores, for each product code that uniquely identifies a product, demand forecast information that indicates the result of demand forecast for that product.

[0045] The demand forecast information includes a forecast date on which the demand forecast was made, a wide-area demand forecast evaluation value, and a local demand forecast evaluation value.

[0046] The wide-area demand forecast evaluation value is an evaluation value that indicates the results of predicting whether wide-area demand is expected in the future. The wide-area demand forecast evaluation value will be described in detail later (see FIG. 9).

[0047] The local demand forecast evaluation value is an evaluation value that indicates the result of predicting whether demand is expected in a specific area in the future. The local demand forecast evaluation value will be described in detail later (see FIG. 10).

[0048] (Functional configuration of the demand forecasting device) This will be explained using Fig. 8. Fig. 8 is a functional block diagram showing an example of the functional configuration of the demand prediction device.

[0049] The control unit 21 of the demand forecasting device 12 deploys the control program 26 in the RAM 24 and runs it to realize, as functional units, a sales increase product detection unit 41, a search count acquisition unit 42, an opinion type analysis unit 43, a demand forecasting unit 46, a forecast result output unit 47, a display control unit 48, an operation control unit 49, and a communication control unit 50, all of which are shown in Fig. 8. Note that some or all of these functions may be realized by dedicated hardware.

[0050] The sales increase product detection unit 41 detects a product (first product) whose sales volume is generally on the rise. The sales increase product detection unit 41 also detects a product (second product) whose sales volume is on the rise for specific users in specific regions. The sales increase product detection unit 41 is an example of a detection unit in the present disclosure. Specifically, the sales increase product detection unit 41 compares the current sales volume of a focused product with the past sales volume (for example, the average sales volume several years ago, one year ago, several weeks ago, or several days ago) to determine whether the sales volume is on the rise.

[0051] The search count acquisition unit 42 acquires the number of searches for a first product from a preset first information source 14. The search count acquisition unit 42 also acquires the number of searches for a second product from a preset second information source 15 that transmits information related to a specific region. The search count acquisition unit 42 also quantifies the degree of increase in the acquired number of searches. The search count acquisition unit 42 is an example of an acquisition unit in the present disclosure.

[0052] The opinion type analysis unit 43 calculates the proportion of positive opinions about the first product from among the opinions received from the first information source 14 or the second information source 15. The opinion type analysis unit 43 also calculates the proportion of positive opinions about the second product from among the opinions received from the second information source. The opinion type analysis unit 43 is an example of a calculation unit in the present disclosure.

[0053] The opinion type analysis unit 43 further includes an opinion extraction unit 44 and an opinion content identification unit 45.

[0054] The opinion extraction unit 44 extracts opinions related to a specific product from among the opinions submitted to the first information source 14. The opinion extraction unit 44 also extracts opinions related to a specific product from among the opinions submitted to the second information source 15. The specific product is a first product that is tending to have a wide increase in sales volume, or a second product that is tending to have an increase in sales volume to specific users in a specific region, as detected by the sales increase product detection unit 41. More specifically, the opinion extraction unit 44 extracts opinions including the name of a specific product from among the opinions submitted to the first information source 14 or the second information source 15. The opinion extraction unit 44 is an example of an extraction unit in the present disclosure.

[0055] The opinion content identification unit 45 identifies whether the opinion is positive, negative, or neither by syntactically analyzing the opinion extracted by the opinion extraction unit 44. More specifically, the opinion content identification unit 45 identifies whether the opinion is positive about a specific product but negative, or neither, by analyzing the dependency structure of the opinion extracted by the opinion extraction unit 44. The opinion content identification unit 45 is an example of an identification unit in the present disclosure.

[0056] The classification process performed by the opinion content classification unit 45 may be performed using a pre-trained model. The pre-trained model is, for example, a model that, in response to input of the content of an opinion and the name of a product of interest, outputs an evaluation value that is an opinion that approves of the product, an evaluation value that is an opinion that disapproves of the product, and an evaluation value that is an opinion that is neither approve nor disapproves of the product.

[0057] The demand forecasting unit 46 determines whether widespread demand for the first product is expected based on the number of searches acquired by the search count acquisition unit 42 and the proportion of positive opinions calculated by the opinion type analysis unit 43. In addition, the demand forecasting unit 46 determines whether widespread demand for the second product is expected based on the number of searches acquired by the search count acquisition unit 42 and the proportion of positive opinions calculated by the opinion type analysis unit 43.

[0058] The prediction result output unit 47 outputs the determination result of the demand prediction unit 46. The prediction result output unit 47 is an example of the output unit in the present disclosure.

[0059] The display control unit 48 generates screen information to be displayed on the display device 34 and controls the display of the display device 34 .

[0060] The operation control unit 49 acquires operation information performed on the operation device 35 and transmits it to the control unit 21 .

[0061] The communication control unit 50 controls communication between the demand prediction device 12 and the POS terminal 18, the POS terminal 19, and the sales management server 16.

[0062] (Explanation of wide-area demand forecasting method) A method for predicting whether widespread demand is expected in the future for a specific product whose sales volume is tending to increase widely will be described using Figure 9. Figure 9 is a diagram illustrating a method by which a demand prediction device predicts whether widespread demand is expected in the future for a specific product. Note that in the following description, predicting whether widespread demand is expected in the future for a specific product will be referred to as wide-area demand forecasting.

[0063] The demand forecasting unit 46 generates a demand forecast evaluation sheet 60 shown in Figure 9 by obtaining and analyzing the degree of increase 64 in the number of searches for the product in the information source 61, the proportion 65 of positive opinions posted, and the degree of increase 66 in the number of purchases by specific users for the product that has been detected by the sales increase product detection unit 41 and is showing a general tendency for sales to increase.

[0064] Whether a product is showing a general tendency for sales to increase is determined by analyzing the contents of the product sales data 27 described above, comparing the current sales volume with past sales volume, and determining, for example, whether the sales volume growth rate exceeds a predetermined threshold value.

[0065] Information source 61 is an item in which first information source 14 and second information source 15 are organized by specific information source. In the example of FIG. 9, various SNS sites 69 are selected as first information source 14. Then, multiple information sources 63 are selected as specific examples of SNS sites 69. Information sources 63 are, for example, image posting sites, video posting sites, message posting sites, personal sites of influencers, etc., and are described as information source #1, information source #2, and information source #3 in FIG. 9.

[0066] 9, various web pages 70 are selected as second information sources 15. A plurality of information sources 63 are selected as specific examples of the web pages 70. The information sources 63 are web pages that contain information about specific regions, and in FIG. 9, are listed as City Hall A and School B.

[0067] The types of information sources 61 are not limited to SNS and web pages, but also include various search service sites and trend analysis sites.

[0068] The degree of search volume increase 64 is an item that quantifies the degree of increase in the number of searches in each information source 63 of the information source 61 for products that are showing a general trend of increasing sales. Specifically, the change in the number of searches is read at predetermined time intervals (every other day, every other month, etc.) from the product search volume data 29 described above. Then, an evaluation value such as 1, 3, 5, etc. is assigned depending on the rate of increase in the number of searches. If the number of searches remains unchanged or is decreasing, the evaluation value is set to 0.

[0069] The proportion of positive opinions 65 is an item indicating the proportion of positive opinions among the contents of opinions posted to each information source 63 included in the first information source 14. Specifically, an evaluation value (0.0 to 1.0) is assigned according to the proportion of positive opinions among the number of posts in a predetermined period from the opinion type data 30 described above.

[0070] The degree of increase in purchase quantity of a specific user 66 is an item that indicates the degree of increase in the purchase quantity of a specific product by a specific user in a specific area. The evaluation value of this item is an item used for local demand forecasting, and will be described in detail later (see FIG. 10).

[0071] Based on the acquired information, the demand forecasting unit 46 calculates a wide-area demand forecast evaluation value 67 and a local demand forecast evaluation value 68. The wide-area demand forecast evaluation value 67 is an evaluation value that indicates whether demand for the product of interest is widely expected. The wide-area demand forecast evaluation value 67 is the product of the degree of increase in the number of searches 64 and the proportion of positive opinions 65.

[0072] The local demand forecast evaluation value 68 is an evaluation item that is applied to products that do not show a general trend of increasing sales volume, and is therefore not applied to Fig. 9. The contents of this item will be described in detail later (see Fig. 10).

[0073] The demand forecasting unit 46 analyzes the calculation result of the wide-area demand forecast evaluation value 67 to determine whether wide-area demand for the product of interest can be expected. For example, the demand forecasting unit 46 calculates a demand forecast evaluation value 71 by adding up all the wide-area demand forecast evaluation values ​​67, and if the demand forecast evaluation value 71 is equal to or greater than a threshold, it determines that wide-area demand for the product of interest can be expected. That is, the demand forecast evaluation value 71 in this case indicates the wide-area demand forecast evaluation value. In the example of FIG. 9, the demand forecast evaluation value 71 obtained by adding up all the wide-area demand forecast evaluation values ​​67 is 4.9. If the threshold for determining that wide-area demand for the product of interest can be expected is set to, for example, 4.5, it is determined that wide-area demand for the product of interest can be expected.

[0074] According to this determination method, the demand forecast evaluation value 71 increases as the number of information sources 63 increases, so the average value of all the wide-area demand forecast evaluation values ​​67 may be used as the demand forecast evaluation value 71 .

[0075] Furthermore, the method of determining whether wide-area demand is expected is not limited to this. For example, it may be determined that wide-area demand is expected for the product of interest when at least one wide-area demand forecast evaluation value 67 among the wide-area demand forecast evaluation values ​​67 for the multiple information sources 63 exceeds a predetermined value.

[0076] (Explanation of local demand forecasting method) A method for predicting whether future demand for a specific product is expected in a specific region will be described using Figure 10. Figure 10 is a diagram illustrating a method by which a demand prediction device predicts whether future demand for a specific product is expected in a specific region. Note that in the following description, predicting whether future demand for a specific product is expected in a specific region will be referred to as local demand prediction.

[0077] 10 is an example of a demand forecast for a product that has been determined by the sales increase product detection unit 41 to have no general trend of increasing sales. The contents of each item on the demand forecast evaluation sheet 72 are the same as those on the demand forecast evaluation sheet 60 (see FIG. 9).

[0078] The degree of search increase 64 shown in Figure 10 indicates that for the particular product of interest, the number of searches from the first information source 14 has not increased, but the number of searches from the second information source 15 has increased.

[0079] The proportion of positive opinions 65 is an item indicating the proportion of positive opinions among the contents of opinions posted to each information source 63 included in the second information source 15. Specifically, an evaluation value (0.0 to 1.0) is assigned according to the proportion of positive opinions among the number of posts in a predetermined period from the opinion type data 30 described above.

[0080] The degree of increase in purchases by a specific user 66 indicates that the number of purchases of a specific product by a specific user in a specific area is increasing rapidly. Note that the degree of increase in purchases by a specific user 66 is assigned an evaluation value of, for example, "1" if the number of purchases has increased compared to one month ago, and "3" if the number of purchases has increased rapidly.

[0081] The wide-area demand forecast evaluation value 67 is not applied to FIG. 10 because it is an evaluation value for products whose sales volume is tending to increase widely.

[0082] The local demand forecast evaluation value 68 is an evaluation value that indicates whether demand for the product of interest can be expected in a specific region. The local demand forecast evaluation value 68 is a value obtained by multiplying the degree of increase in the number of searches 64 and the proportion of positive opinions 65.

[0083] The demand forecasting unit 46 analyzes the calculation result of the local demand forecast evaluation value 68 and determines whether demand for the product of interest can be expected in the specific region. For example, the demand forecasting unit 46 calculates a demand forecast evaluation value 71 by adding all the local demand forecast evaluation values ​​68 together and further adding the purchase increase degree 66 of the specific user. If the demand forecast evaluation value 71 is equal to or greater than a threshold, it is determined that demand for the product of interest can be expected in the specific region. That is, the demand forecast evaluation value 71 in this case indicates a local demand forecast evaluation value. In the example of FIG. 10 , the demand forecast evaluation value 71 is 5.0, obtained by adding the evaluation value 2.0 obtained by adding all the local demand forecast evaluation values ​​68 together and the evaluation value 3.0 of the purchase increase degree 66 of the specific user. If the threshold for determining that demand for the product of interest can be expected in the specific region is set to, for example, 3.5, it is determined that local demand for the product of interest can be expected. Furthermore, when it is determined that wide-area demand is expected, the demand forecasting unit 46 generates a wide-area demand forecast screen 90 (see FIG. 12) showing the determination result. Furthermore, when it is determined that local demand is expected, the demand forecasting unit 46 generates a local demand forecast screen 98 (see FIG. 13) showing the determination result.

[0084] According to this determination method, the demand forecast evaluation value 71 increases as the number of information sources 63 increases, so the demand forecast evaluation value 71 may be calculated by adding the average value of all local demand forecast evaluation values ​​68 to the degree of increase in purchases 66 of a specific user.

[0085] Furthermore, the method for determining whether local demand can be expected is not limited to this. For example, it may be determined that local demand for the product of interest can be expected when the sum of at least one local demand forecast evaluation value 68 among the local demand forecast evaluation values ​​68 for the multiple information sources 63 and the degree of increase in purchase quantity 66 exceeds a predetermined value.

[0086] (Example of demand forecast result output) An example of the demand forecast result output by the demand forecasting device 12 will be described using Fig. 11, Fig. 12, and Fig. 13. Fig. 11 is a diagram showing an example of the forecast result output by the demand forecasting device on a product ordering screen. Fig. 12 is a diagram showing an example of the wide-area demand forecast result output by the demand forecasting device. Fig. 13 is a diagram showing an example of the local demand forecast result output by the demand forecasting device.

[0087] A product ordering screen 80 displayed on the display device 34 of the demand forecasting device 12 displays a product display field 81, an order quantity setting key 82, a wide-area demand increase marker 83, and a local demand increase marker 84.

[0088] The product display field 81 displays information indicating the product to be ordered, such as the product name and appearance of the product.

[0089] The order quantity setting key 82 is a button for specifying the number of items to be ordered. When the + key is pressed, the order quantity increases according to the number of times it is pressed. Also, when the - key is pressed, the order quantity decreases according to the number of times it is pressed. Although not shown, the specified order quantity is superimposed on the product display field 81, for example.

[0090] The wide-area demand increase marker 83 is superimposed on the product display field 81 of the product for which the demand forecasting device 12 predicts that the wide-area demand will increase.

[0091] The local demand increase marker 84 is superimposed on the product display field 81 of the product for which the demand forecasting device 12 predicts that the local demand will increase.

[0092] The example in FIG. 11 shows that the wide-area demand for product B is predicted to increase, and the local demand for product F is predicted to increase.

[0093] Although not shown in Figure 11, the product order screen 80 also displays various operation buttons, such as a page forward button for moving to a page displaying another product, an order instruction button for instructing an order for the set order quantity, and a cancel button for canceling the set order quantity.

[0094] The wide-area demand increase marker 83 and the local demand increase marker 84 also serve as operation buttons. When the wide-area demand increase marker 83 is pressed, a wide-area demand forecast screen 90 for product B, as shown in FIG. 12, is displayed on the display device 34. Furthermore, when the local demand increase marker 84 is pressed, a local demand forecast screen 98 for product F, as shown in FIG.

[0095] The wide-area demand forecast screen 90 is a screen that displays the results of a wide-area demand forecast performed by the demand forecasting device 12 for a product (first product) that is showing a tendency for sales to increase widely, and the reasons why the demand forecasting device 12 has determined that there is likely to be widespread demand for the product.

[0096] As shown in Figure 12, the wide-area demand forecast screen 90 includes a reason 91 for determining that there is widespread demand for the product, a sales volume trend 92 for the product, a trend 93 for the number of searches related to the product, a breakdown of opinion types 94, and a wide-area demand forecast evaluation value 67.

[0097] The reason for determination 91 is a statement that briefly summarizes the reason why the demand forecasting unit 46 has determined that the focused product is likely to have widespread demand. In the example of Fig. 12, a statement is displayed that indicates that the number of searches for product B has recently increased and that sales have also increased.

[0098] The sales volume transition 92 is a graph showing the transition of the sales volume of products that are generally on the rise in sales volume, as detected by the sales increase product detection unit 41.

[0099] The search count transition 93 is a graph showing the transition of the search count for each information source 63 calculated by the search count acquisition unit 42 .

[0100] The opinion type breakdown 94 is a graph showing the breakdown of opinion types (positive opinions, negative opinions, neutral opinions) calculated by the opinion type analysis unit 43 for each information source 63 where opinions can be posted. Note that the opinion type breakdown 94 only needs to include at least the proportion of positive opinions. In other words, the opinion type breakdown 94 may be a numerical value indicating the proportion of positive opinions.

[0101] The wide-area demand forecast evaluation value 67 is an evaluation value calculated by the demand forecasting unit 46, which indicates whether or not wide-area demand for the product in question can be expected. Note that it is desirable to display the wide-area demand forecast evaluation value 67 together with a threshold value for determining whether wide-area demand can be expected, as shown in FIG. 12 .

[0102] The local demand forecast screen 98 is a screen that includes the results of a local demand forecast performed by the demand forecasting device 12 for a product (second product) that tends to see an increase in sales to specific users in a specific area, and the reasons why the demand forecasting device 12 has determined that there is widespread demand for the product.

[0103] As shown in Figure 13, the local demand forecast screen 98 includes a reason 91 for determining that there is widespread demand for the product, a sales volume trend 92 for the product, a trend 93 for the number of searches related to the product, a breakdown of opinion types 94, and a local demand forecast evaluation value 68.

[0104] The reason for determination 91 is a statement that briefly summarizes the reason why the demand forecasting unit 46 has determined that there is likely to be demand for the product of interest in a specific region. In the example of Fig. 13, a statement is displayed that indicates that there has been an increase in the number of searches for product F in a specific region recently and that there has been an increase in sales in the specific region.

[0105] The product sales volume trend 92, the product search volume trend 93, and the opinion type breakdown 94 are as described with reference to FIG.

[0106] The local demand forecast evaluation value 68 is an evaluation value calculated by the demand forecasting unit 46 that indicates whether demand for the product of interest can be expected in a specific region. It is desirable to display the local demand forecast evaluation value 68 together with a threshold value for determining whether demand can be expected in a specific region, as shown in Fig. 13. The threshold value for determining whether demand can be expected widely, as shown in Fig. 12, and the threshold value for determining whether demand can be expected in a specific region, as shown in Fig. 13, are values ​​that are set independently based on past demand forecast results, etc., and therefore do not necessarily match, as shown in Figs. 12 and 13.

[0107] (Processing flow performed by the demand forecasting device) The flow of processing performed by the demand prediction device 12 will be described with reference to Fig. 14. Fig. 14 is a flowchart showing an example of the flow of processing performed by the demand prediction device.

[0108] The sales increase product detection unit 41 determines whether there is a product whose sales volume is on the rise (step S11). If it is determined that there is a product whose sales volume is on the rise (step S11: Yes), the process proceeds to step S12. On the other hand, if it is not determined that there is a product whose sales volume is on the rise (step S11: No), the process proceeds to step S16.

[0109] If it is determined in step S11 that there is a product whose sales volume is on the rise, the demand forecasting unit 46 performs wide-area demand forecasting processing to determine whether there is wide demand for the product whose sales volume is on the rise (step S12). The flow of the wide-area demand forecasting processing will be described later (see FIG. 15).

[0110] Next, the demand forecasting unit 46 determines whether the calculated wide-area demand forecast evaluation value 67 is equal to or greater than a threshold value (step S13). If it is determined that the wide-area demand forecast evaluation value 67 is equal to or greater than the threshold value (step S13: Yes), the process proceeds to step S14. On the other hand, if it is not determined that the wide-area demand forecast evaluation value 67 is equal to or greater than the threshold value (step S13: No), the process proceeds to step S15.

[0111] If it is determined in step S13 that the wide-area demand forecast evaluation value 67 is equal to or greater than the threshold value, the demand forecasting unit 46 determines that the product of interest has wide-area demand (step S14), and then proceeds to step S21.

[0112] If it is determined in step S13 that the wide-area demand forecast evaluation value 67 is not equal to or greater than the threshold value, the demand forecasting unit 46 determines that there is no demand for the product of interest (step S15). After that, the demand forecasting device 12 ends the processing of FIG. 14.

[0113] Returning to step S11, if it is determined in step S11 that there is no product whose sales volume is on the rise, the sales increase product detection unit 41 determines whether there is a product whose sales volume is on the rise for the specific user (step S16). If it is determined that there is a product whose sales volume is on the rise for the specific user (step S16: Yes), the process proceeds to step S17. On the other hand, if it is not determined that there is a product whose sales volume is on the rise for the specific user (step S16: No), the process proceeds to step S20.

[0114] In step S16, if it is determined that there is a product whose sales volume is on the rise for a specific user, the demand forecasting unit 46 performs local demand forecasting processing to determine whether there is a demand for the product whose sales volume is on the rise in the specific region (step S17). The flow of the local demand forecasting processing will be described later (see FIG. 16).

[0115] Next, the demand forecasting unit 46 determines whether the calculated local demand forecast evaluation value 68 is equal to or greater than a threshold value (step S18). If it is determined that the local demand forecast evaluation value 68 is equal to or greater than the threshold value (step S18: Yes), the process proceeds to step S19. On the other hand, if it is not determined that the local demand forecast evaluation value 68 is equal to or greater than the threshold value (step S18: No), the process proceeds to step S20.

[0116] If it is determined in step S18 that the local demand forecast evaluation value 68 is equal to or greater than the threshold value, the demand forecasting unit 46 determines that the product of interest has local demand (step S19), and then proceeds to step S21.

[0117] If it is determined in step S18 that the local demand forecast evaluation value 68 is not equal to or greater than the threshold value, the demand forecasting unit 46 determines that there is no demand for the product of interest (step S20). After that, the demand forecasting device 12 ends the processing of FIG. 14.

[0118] Following step S14 or step S19, the demand forecasting unit 46 generates a wide-area demand forecast screen 90 or a local demand forecast screen 98, which is output information indicating the determination result (step S21).

[0119] The demand forecasting unit 46 determines whether an instruction to display output information indicating the determination result has been issued by pressing the wide-area demand increase marker 83, which indicates that there is wide-area demand, or the local demand increase marker 84, which indicates that there is local demand, on the product ordering screen 80 (step S22). If it is determined that an instruction to display output information has been issued (step S22: Yes), the process proceeds to step S23. On the other hand, if it is not determined that an instruction to display output information has been issued (step S22: No), the demand forecasting device 12 ends the process of FIG. 14.

[0120] (Wide-area demand forecast processing flow) The flow of the wide-area demand forecasting process will be described with reference to Fig. 15. Fig. 15 is a flowchart showing an example of the flow of the wide-area demand forecasting process.

[0121] The search count acquisition unit 42 acquires the number of searches for a product determined to have an increasing sales trend in step S11 (see FIG. 14) for one of the preset first information sources 14. Then, the search count acquisition unit 42 calculates the search count increase degree 64 (see FIG. 9) (step S31).

[0122] The opinion type analysis unit 43 analyzes whether the posts made to the first information source 14 about the product of interest are positive opinions, negative opinions, or neutral opinions (step S32).

[0123] Next, the opinion type analysis unit 43 calculates the proportion 65 of positive opinions about the product of interest (see FIG. 9) (step S33).

[0124] The demand forecasting unit 46 calculates the wide-area demand forecast evaluation value 67 (see FIG. 9) for the product of interest for the first information source 14 of interest (step S34).

[0125] The demand forecasting unit 46 determines whether all of the first information sources 14 have been evaluated (step S35). If it is determined that all of the first information sources 14 have been evaluated (step S35: Yes), the process proceeds to step S36. On the other hand, if it is determined that all of the first information sources 14 have not been evaluated (step S35: No), the process returns to step S31, and a different first information source 14 is analyzed.

[0126] If it is determined in step S35 that all of the first information sources 14 have been evaluated, the demand forecasting unit 46 determines a demand forecast evaluation value 71 (see FIG. 9) by adding the wide-area demand forecast evaluation values ​​67 calculated for all of the first information sources 14 (step S36). Thereafter, the process returns to the main routine (FIG. 14).

[0127] (Local demand forecast processing flow) The flow of the local demand forecasting process will be described with reference to Fig. 16. Fig. 16 is a flowchart showing an example of the flow of the local demand forecasting process.

[0128] The search count acquisition unit 42 acquires the number of searches for a product determined to have an increasing sales trend in step S16 (see FIG. 14) for one of the preset second information sources 15. Then, the search count acquisition unit 42 calculates the search count increase degree 64 (see FIG. 10) (step S41).

[0129] The opinion type analysis unit 43 analyzes whether the posts made to the second information source 15 about the product of interest are positive opinions, negative opinions, or neutral opinions (step S42).

[0130] Next, the opinion type analysis unit 43 calculates the proportion 65 of positive opinions about the product of interest (see FIG. 10) (step S43).

[0131] The demand forecasting unit 46 calculates a local demand forecast evaluation value 68 (see FIG. 10) for the product of interest for the second information source 15 of interest (step S44).

[0132] The demand forecasting unit 46 determines whether all second information sources 15 have been evaluated (step S45). If it is determined that all second information sources 15 have been evaluated (step S45: Yes), the process proceeds to step S46. On the other hand, if it is determined that not all second information sources 15 have been evaluated (step S45: No), the process returns to step S41, and a different second information source 15 is analyzed.

[0133] If it is determined in step S45 that all second information sources 15 have been evaluated, the demand forecasting unit 46 determines a demand forecast evaluation value 71 (see FIG. 10) by adding the local demand forecast evaluation values ​​68 calculated for all second information sources 15 and the purchase quantity increase degree 66 of the specific user (step S46). Then, the process returns to the main routine (FIG. 14).

[0134] (Effects of the embodiment) As described above, the demand forecasting device 12 (information processing device) of this embodiment includes a sales increase product detection unit 41 (detection unit) that detects a first product (product) whose sales volume is tending to increase widely, a search count acquisition unit 42 (acquisition unit) that acquires the number of searches for the first product from a preset first information source 14, an opinion type analysis unit 43 (calculation unit) that calculates the proportion of positive opinions about the first product from among the opinions submitted to the first information source 14, a demand forecasting unit 46 that determines whether widespread demand for the first product is expected based on the number of searches acquired by the search count acquisition unit 42 and the proportion of positive opinions calculated by the opinion type analysis unit 43, and a forecast result output unit 47 (output unit) that outputs the determination result of the demand forecasting unit 46. Therefore, products whose demand is expected to increase widely in the future can be predicted early.

[0135] Furthermore, in the demand forecasting device 12 (information processing device) of this embodiment, the sales increase product detection unit 41 (detection unit) further detects a second product (product) whose sales volume to specific users in a specific region tends to increase, the search count acquisition unit 42 (acquisition unit) further acquires the number of searches for the second product from a second information source 15 that is set in advance and transmits information related to the specific region, the opinion type analysis unit 43 (calculation unit) further calculates the proportion of positive opinions about the second product from the opinions received by the second information source 15, and the demand forecasting unit 46 further determines whether demand for the second product is expected in the specific region based on the number of searches acquired by the search count acquisition unit 42 and the proportion of positive opinions calculated by the opinion type analysis unit 43. Therefore, it is possible to early predict products whose demand is expected to increase in the future in a specific region.

[0136] Furthermore, in the demand forecasting device 12 (information processing device) of this embodiment, the forecast result output unit (output unit) outputs a reason 91 for the determination made by the demand forecasting unit 46, a sales volume trend 92, a search volume trend 93, and an opinion type breakdown 94 including the proportion of positive opinions. Therefore, it is possible to inform the user of the reason why it was determined that demand is likely to be expected.

[0137] Furthermore, in the demand forecasting device 12 (information processing device) of this embodiment, the forecast result output unit 47 (output unit) outputs the determination result as to whether demand for the product is expected, in association with the product, to the product ordering screen 80. Therefore, the person in charge of instructing the number of products to be ordered can easily recognize products for which demand is expected to increase, thereby improving the operability of operations related to ordering work.

[0138] Furthermore, in the demand forecasting device 12 (information processing device) of this embodiment, the opinion type analysis unit 43 (calculation unit) includes an opinion extraction unit 44 (extraction unit) that extracts opinions related to a product, and an opinion content identification unit 45 (identification unit) that identifies whether the opinions are positive, negative, or neither by syntactically analyzing the opinions extracted by the opinion extraction unit 44. Therefore, it is possible to reliably analyze whether opinions about a product submitted to an information source are positive.

[0139] Although the embodiments of the present invention have been described above, these embodiments are merely examples and are not intended to limit the scope of the invention. This novel embodiment can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0140] 10 Demand forecasting system 12 Demand forecasting device (information processing device) 14 Primary Source 15 Secondary Sources 16 Sales management server 18, 19 POS terminals 27 Product sales data 28 Specific user purchase data 29 Product search volume data 30 Opinion type data 31 Demand forecast data 32 Knowledge data for syntactic analysis 41 Sales Increase Item Detection Unit (Detection Unit) 42 Search number acquisition unit (acquisition unit) 43 Opinion type analysis unit (calculation unit) 44 Opinion extraction part (extraction part) 45 Opinion content identification unit (identification unit) 46 Demand Forecasting Department 47 Prediction result output unit (output unit) 60, 72 Demand forecast evaluation sheet 67 Wide-area demand forecast evaluation value 68 Local demand forecast evaluation value 80 Product ordering screen 90 Wide-area demand forecast screen 91 Reason for judgment 92 Sales Trends 93 Search Count Trends 94 Breakdown of Opinion Types 98 Local demand forecast screen N Network [Prior art documents] [Patent documents]

[0141] [Patent Document 1] Japanese Patent Application Laid-Open No. 2018-160068

Claims

1. A detection unit that compares the current sales volume of a product with the past sales volume and detects products whose sales volume is generally on the rise based on whether the sales volume growth rate exceeds a predetermined threshold; an acquisition unit that acquires the number of searches for the product from a first information source that is set in advance and is expected to be widely accessed nationwide; a calculation unit that calculates a ratio of positive opinions about the product from among the opinions submitted to the first information source; a demand forecasting unit that determines whether or not there is a wide demand for the product based on whether or not a value obtained by adding, for each of the first information sources, the multiplication of the degree of increase in the number of searches acquired by the acquisition unit and the proportion of positive opinions calculated by the calculation unit is equal to or greater than a threshold value; and an output unit that outputs a determination result of the demand forecasting unit; An information processing device comprising:

2. The detection unit further detects products whose sales volume to specific users in specific areas tends to increase, The acquisition unit further acquires the number of searches for the product from a second information source that is set in advance and is expected to be accessed frequently from the specific region, The calculation unit further calculates a ratio of positive opinions about the product from among the opinions submitted to the second information source, The demand forecasting unit further determines whether demand for the product is expected in the specific region based on whether a value obtained by adding, for each of the second information sources, a multiplication value of the degree of increase in the number of searches acquired by the acquisition unit and the proportion of positive opinions calculated by the calculation unit, and a sum of the sum and a degree of increase in the number of purchases of the product by specific users in the specific region is equal to or greater than a threshold value. The information processing device according to claim 1 .

3. The output unit outputs the reason for the determination by the demand forecasting unit, the change in the number of sales, the change in the number of searches, and the ratio of positive opinions.

3. The information processing device according to claim 1.

4. The output unit outputting a determination result as to whether demand for the product is expected, in association with the product, on an ordering screen for the product; 3. The information processing device according to claim 1.

5. The calculation unit an extraction unit that extracts opinions related to the product; and a discrimination unit that discriminates whether the opinion is positive, negative, or neither by syntactically analyzing the opinion extracted by the extraction unit.

3. The information processing device according to claim 1.

6. Computer, a detection unit that compares the current sales volume of a product with the past sales volume and detects products whose sales volume is on the rise based on whether the sales volume increase rate exceeds a predetermined threshold; an acquisition unit that acquires the number of searches for the product from a first information source that is set in advance and is expected to be widely accessed nationwide; a calculation unit that calculates a ratio of positive opinions about the product from among the opinions submitted to the first information source; a demand forecasting unit that determines whether or not there is a wide demand for the product based on whether or not a value obtained by adding, for each of the first information sources, the multiplication of the degree of increase in the number of searches acquired by the acquisition unit and the proportion of positive opinions calculated by the calculation unit is equal to or greater than a threshold value; and an output unit that outputs a determination result of the demand forecasting unit; A program that makes it work.

Images