Information processing device, information processing method, and recording medium

Abstract

An information processing device according to the present disclosure comprises an acquisition unit, an analysis unit, and a display unit. The acquisition unit acquires information regarding a product. The analysis unit analyzes a factor indicating a reason why the product sold on the basis of the information acquired by the acquisition unit. The display unit displays the result of the analysis made by the analysis unit.

Description

Information Processing Apparatus, Information Processing Method, and Recording Medium

[0001] This disclosure relates to an information processing apparatus, an information processing method, and a recording medium.

[0002] A method of comprehensively analyzing a situation from various perspectives in consideration of the scale of customers and markets and the features and shares of the company's own products is known. For example, a method of performing analysis based on sales performance information and outputting an evaluation result from the viewpoint of the market share and market growth rate of a certain product is known (Japanese Patent Application Laid-Open No. 2003-271803, etc.).

[0003] Japanese Patent Application Laid-Open No. 2003-271803

[0004] In various operations such as product development, analysis of products is required. However, there has been a problem that product analysis takes time and cost and places a heavy burden on workers.

[0005] An example of the object of this disclosure is to develop a technology related to product analysis.

[0006] According to one aspect of this disclosure, there is provided an information processing apparatus including: acquisition means for acquiring information related to a product; analysis means for analyzing factors indicating reasons why the product has sold based on the information; and display means for displaying the result of the analysis.

[0007] Further, according to one aspect of this disclosure, there is provided an information processing method in which a computer acquires information related to a product, analyzes factors indicating reasons why the product has sold based on the information, and displays the result of the analysis.

[0008] Further, according to one aspect of this disclosure, there is provided a recording medium recording a program for causing a computer to execute an acquisition step of acquiring information related to a product, an analysis step of analyzing factors indicating reasons why the product has sold based on the information, and a display step of displaying the result of the analysis.

[0009] Figure 1 is a diagram showing an example of a functional block diagram of an information processing device. Figure 2 is a flowchart showing an example of the processing flow of an information processing device. Figure 3 is a diagram showing an example of the hardware configuration of an information processing device. Figure 4 is a diagram showing another example of a functional block diagram of an information processing device. Figure 5 is a flowchart showing another example of the processing flow of an information processing device. Figure 6 is a diagram showing another example of a functional block diagram of an information processing device. Figure 7 is a flowchart showing another example of the processing flow of an information processing device. Figure 8 is a diagram showing another example of a functional block diagram of an information processing device. Figure 9 is a flowchart showing another example of the processing flow of an information processing device. Figure 10 is a flowchart showing another example of the processing flow of an information processing device. Figure 11 is a diagram showing another example of a functional block diagram of an information processing device. Figure 12 is a flowchart showing another example of the processing flow of an information processing device. Figure 13 is a diagram showing an example of a screen provided by an information processing device. Figure 14 is a diagram showing another example of a screen provided by an information processing device.

[0010] The principles of this disclosure will be explained with reference to several exemplary embodiments. These embodiments are provided for illustrative purposes only and should be understood as aiding those skilled in the art to understand and implement this disclosure without implying any limitation on the scope of this disclosure. The disclosures described in this specification may be implemented in various ways other than those described below.

[0011] In the following description and claims, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those generally understood by those skilled in the art to which this disclosure belongs.

[0012] Embodiments of this disclosure will be described below with reference to the drawings. Each drawing is merely illustrative for illustrating one or more embodiments. Each drawing may be associated not only with one specific embodiment but also with one or more other embodiments. As those skilled in the art will understand, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings, for example, to create embodiments not explicitly shown or described. Not all features or steps shown in any one drawing are necessarily required to illustrate an exemplary embodiment, and some features or steps may be omitted. The order of steps shown in any of the drawings may be changed as appropriate.

[0013] (Embodiment 1) Figure 1 is a functional block diagram showing an overview of the information processing device 10. Figure 2 is a flowchart showing an example of the processing flow executed by the information processing device 10.

[0014] As shown in Figure 1, the information processing device 10 includes an acquisition unit 110, an analysis unit 120, and a display unit 130. These functional units execute the processes shown in the flowchart of Figure 2.

[0015] In S110, the acquisition unit 110 acquires information about the product. In S120, the analysis unit 120 analyzes the factors that indicate why the product sold, based on the information acquired in S110. In S130, the display unit 130 displays the results of the analysis in S120.

[0016] Thus, the information processing device 10 has the characteristic of acquiring information about a product, analyzing the product based on that information, and displaying the results of the analysis. By using this distinctive information processing device 10, workers can easily acquire the desired information without spending an unnecessarily large amount of time and cost on product analysis. As a result, workers can allocate their time and resources to other tasks. Furthermore, workers can proceed with their work more efficiently.

[0017] Furthermore, the information processing device 10 has the characteristic of acquiring information about a product, analyzing the "factors indicating why the product sold" based on that information, and displaying the results of that analysis. The "factors indicating why the product sold" are items that greatly influence product sales and are of high importance in product development work. The information processing device 10 can analyze such characteristic items and provide them to the operator. By using this characteristic information processing device 10, the operator can grasp the "factors indicating why the product sold," which are of high importance in product development work.

[0018] Product analysis is performed, for example, in product development. In product development, it is important to analyze data from multiple sources and consider hypotheses for new products. However, the analysis and hypothesis testing process is time-consuming and costly, placing a heavy burden on workers. To address this issue, as described above, workers can use the information processing device 10 to perform product analysis without unnecessarily spending a lot of time and money on analysis. Based on the results of this analysis, workers can then consider hypotheses for new products. In this way, by using the information processing device 10, workers can efficiently advance product development.

[0019] (Embodiment 2) Embodiment 2 will now be described.

[0020] [Explanation of the premise] In new product development, it is important to analyze the factors that led to the success of competitors' products. Until now, information on competitors has been limited, and this has mainly been done by purchasing POS (point of sales) data and having people analyze it. However, there has been a challenge in that it is extremely difficult to analyze the vast and diverse data and arrive at factor analysis. Therefore, the purpose of this embodiment is to enable food manufacturers and other companies developing new products to efficiently perform factor analysis of competitors' products.

[0021] [Description of Configuration] The outline of the information processing device 10 will be described with reference to Figure 1. Figure 1 is a block diagram showing an example of the information processing device 10 in this embodiment. The information processing device 10 comprises an acquisition unit 110, an analysis unit 120, and a display unit 130.

[0022] The acquisition unit 110 acquires information about products. Products may be food, daily necessities, general merchandise, clothing, or other items. For example, the acquisition unit 110 acquires information related to product development. Information related to product development may include consumer feedback, market trends, competitor product information, and sales data. The acquisition unit 110 may automatically acquire information from the internet, databases, sensor data, etc., or it may acquire it through user input or other means. Automatic acquisition can be achieved using widely known technologies. Specifically, the acquisition unit 110 may acquire the above information from social media data analysis, product reviews on e-commerce (EC) sites, user behavior data from IoT (Internet of Things) devices, market research company reports, etc.

[0023] For example, the acquisition unit 110 may collect user posts related to specific keywords from social media as consumer feedback, or acquire product reviews (ratings (number of stars), comments, etc.) from product pages on e-commerce sites. Specific keywords may be set by the product developer or may be pre-set.

[0024] In addition, the acquisition unit 110 may acquire market research reports from market research companies that show rankings of popular products, etc., as information indicating market trends. In addition, the acquisition unit 110 may acquire information indicating social media trends, etc., as information indicating market trends. In addition, the acquisition unit 110 may acquire product information (product name, sales price, product image, product function, etc.) and store information (store address, etc.) disclosed on e-commerce sites or the homepages of competing product vendors, etc., as competitive product information. In addition, the acquisition unit 110 may acquire market research reports from market research companies that show the sales status of products, etc., as sales data. Note that the product information exemplified here is just an example and is not limited to these. In the following explanation, other "product information" will be exemplified as appropriate.

[0025] The analysis unit 120 performs a factor analysis of the product based on the information collected by the acquisition unit 110. Here, factors refer to explanatory variables that explain why competitors' products sold well. All data acquired by the acquisition unit 110, such as store location, product name, and reviews, are included as explanatory variables. The analysis may include data cleaning, filtering, and trend analysis using statistical methods. In addition, consumer needs and market trends may be extracted using text mining or machine learning algorithms.

[0026] The analysis unit 120 may use, for example, XAI (eExplainable Artificial Intelligence) as a method for identifying specific factors. By using XAI, it is possible to provide an interpretable explanation for the prediction results of the machine learning model and clarify which factors are influencing product sales.

[0027] The analysis unit 120 may output an explanation of how the model arrived at a particular prediction or decision as a result of the analysis. An example of such an explanation is shown below. As in this example, the analysis unit 120 can explain the decision process and rationale of the AI ​​model in a way that is understandable to humans. This increases the reliability of the analysis results and enables decision-makers to appropriately interpret and utilize the results.

[0028] "Product A has high sales projections because its price range is appropriate, its design matches consumer preferences, and its functionality is superior to competing products. Specifically, its price is 10% lower than the market average, 90% of consumer reviews regarding its design are positive, and the inclusion of unique feature X is a strength that competing products lack."

[0029] Furthermore, the analysis unit 120 may perform the analysis using at least one of the following: automatic feature engineering, large-scale language models, and genetic algorithms. Automatic feature engineering can automatically generate new features from primitive features and identify important factors that influence product sales from among them. By utilizing large-scale language models, latent factors can be extracted from text data, and factor analysis that takes complex contexts into account can be performed. By using genetic algorithms, a large number of combinations of latent factors can be efficiently explored, and the optimal combination of factors can be found. These methods are examples, and other methods may be used. Furthermore, the analysis unit 120 may use a combination of multiple methods.

[0030] As an explainable AI (XAI), the analysis unit 120 may use methods such as LIME (Local Interpretable Model-agnostic Explanations) or SHAP (SHapley Additive exPlanations). By using these methods, it is possible to visualize and explain in an interpretable way which features have an influence on the prediction results of a machine learning model and to what extent. For example, in a sales forecasting model for a certain product, it is possible to quantitatively show to what extent factors such as "price," "design," and "functionality" contribute to sales.

[0031] As a concrete example of automated feature engineering, the analysis unit 120 can use algorithms such as DFS (Deep Feature Synthesis) and AutoFeat. In DFS, various mathematical operations (addition, subtraction, multiplication, division, etc.) are applied to the original features to generate new features. This makes it possible to automatically create complex features that humans would not think of, and to discover potential factors that affect product sales.

[0032] Examples of large-scale language models include GPT-3 (Generative Pretrained Transformer 3) and BERT (Bidirectional Encoder Representations from Transformers). The analysis unit 120 can use such large-scale language models to extract consumers' potential needs and product evaluation points from product reviews and social media posts. For example, the analysis unit 120 can extract factors such as "ease of use," "convenience," and "lightweight" from a review such as "This product is easy to use and convenient, but it would be even better if it were a little lighter," and analyze their importance.

[0033] As a concrete example of a genetic algorithm, the analysis unit 120 could represent various product characteristics (price, design, function, etc.) as genes, and set sales and customer satisfaction (average number of stars in product reviews, etc.) as fitness indicators. By repeatedly running this algorithm, it is possible to search for combinations of characteristics that bring high sales and customer satisfaction, and to find the optimal direction for product design.

[0034] Furthermore, the analysis unit 120 may perform factor analysis by combining these methods. For example, the analysis unit 120 may perform factor analysis based on both quantitative and qualitative data. Quantitative data may include the number of products sold, price, inventory levels, customer data (age, gender, purchase frequency, average purchase amount, etc.), website data (page views, time spent on site, click-through rate, etc.), and financial data. Qualitative data may include product specifications, product reviews posted on e-commerce sites and app stores, social media posts about products and companies, customer support records (inquiries, complaints, suggestions, etc.), interview results (opinions on product usage experience, etc.), and survey responses (requests for product improvements and new features, etc.).

[0035] The analysis unit 120 may employ a combined approach, for example, as follows. In this example, the analysis unit 120 generates features from both quantitative and qualitative data using automatic feature design. The analysis unit 120 also analyzes at least one of the generated features and qualitative data using a large-scale language model. Finally, the analysis unit 120 optimizes these generation and analysis results using a genetic algorithm and explains them in a form that can be interpreted by XAI.

[0036] Specifically, the analysis unit 120 first generates features from sales data, product specifications, and prices using automated feature design. At this stage, it identifies fundamental factors that may influence product sales. Next, the analysis unit 120 analyzes product review text using a large-scale language model and extracts abstract factors such as "ease of use" and "design" from consumer feedback. Simultaneously, the analysis unit 120 also evaluates the importance of the previously generated features using the large-scale language model.

[0037] The extracted abstract factors are then used again as input for automatic feature engineering. For example, if the factor "ease of use" is extracted, the analysis unit 120 generates specific indicators related to this "ease of use" (such as the number of operating steps or the user interface evaluation score) through automatic feature engineering. This allows abstract concepts to be converted into more specific and measurable indicators.

[0038] Subsequently, the analysis unit 120 treats all factors and features obtained in the previous processes as genes and searches for the optimal combination using a genetic algorithm. In this process, the analysis unit 120 identifies the combination of factors that has the greatest impact on product sales. Finally, the analysis unit 120 uses XAI to explain in an interpretable way how the identified combination of factors affects product sales. Alternatively, the analysis unit 120 may also explain the optimal combination of factors identified by the genetic algorithm in natural language using a large-scale language model. This makes it possible to present the analysis results in a way that is easy for stakeholders without a technical background to understand.

[0039] The analysis unit 120 employs multiple analytical methods, enabling comprehensive factor analysis that considers both quantitative and qualitative data. Using the results of such factor analysis, more effective product development strategies can be formulated. Furthermore, by iteratively applying this process, a flexible analytical system can be built that can quickly respond to market changes and new trends.

[0040] The display unit 130 displays the factors contributing to the success of competitors' products, as analyzed by the analysis unit 120, in a format that is easy for the user to understand. Display formats include graphs, charts, text explanations, 3D models, and dashboards. Furthermore, a display method that allows for the comparison of multiple factors may be employed to support decision-making. For example, the display unit 130 may display the importance of factors using a bar chart or show the correlation between factors using a correlation matrix. Additionally, the display unit 130 may employ dynamic display methods, such as representing changes in factors over time using animation.

[0041] [Description of Operations] FIG. 2 is a flowchart showing an example of the operations of the information processing apparatus 10. An example of a series of processes of the information processing apparatus 10 will be described with reference to FIG. 2.

[0042] The acquisition unit 110 collects information related to product development (step S110). This information may include consumer feedback, market trends, competitive product information, sales data, and the like. Specifically, the acquisition unit 110 may acquire information from social media data, product reviews on e-commerce sites, user behavior data from IoT devices, reports from market research companies, and the like.

[0043] The analysis unit 120 performs factor analysis of the product based on the information related to product development acquired by the acquisition unit 110 (step S120). As analysis methods, techniques such as XAI, automatic feature design, large language models, and genetic algorithms may be used. Also, analysis of consumer reviews using text mining techniques, time series analysis of market trends using statistical methods, and the like may be performed. The analysis unit 120 may use these methods alone or in combination. Note that the analysis unit 120 may perform preprocessing of the data before analysis. Examples of preprocessing include deletion of duplicate data, completion of missing values, handling of outliers, and noise removal of text data (such as deletion of HTML (Hyper Text Markup Language) tags).

[0044] The display unit 130 displays the analysis results analyzed by the analysis unit 120 (step S130). Display methods include graphs, charts, text descriptions, 3D models, dashboards, and the like. Also, the display unit 130 may display information on a display, project information with a projector, or display information by other methods. As an example of display, for example, the display unit 130 may display the importance of factors in a bar chart or show the correlation relationship between factors in a correlation matrix. Also, the display unit 130 may adopt a dynamic display method such as expressing the change of factors over time in an animation.

[0045] [Hardware Configuration] An example of the hardware configuration of the information processing device 10 is described below. Each functional unit of the information processing device 10 is realized by any combination of hardware and software. It will be understood by those skilled in the art that there are various modifications to the implementation method and the device. The software includes programs that are pre-installed at the time of shipment of the device, as well as programs downloaded from recording media such as CDs (Compact Discs) or from servers on the Internet.

[0046] Figure 3 shows an example of the hardware configuration of the information processing device 10. In the example in Figure 3, the information processing device 10 (computer 100) includes a processor 101, memory 102, and a communication interface 103. These parts may be connected by a bus or the like. The memory 102 stores at least a portion of the program 104. The communication interface 103 includes an interface necessary for communication with other network elements.

[0047] When program 104 is executed in cooperation with the processor 101 and memory 102, etc., the computer 100 performs at least some of the processing of embodiments of this disclosure. Memory 102 may be of any type. Memory 102 may, in non-limited examples, be a non-temporary computer-readable storage medium. Memory 102 may also be implemented using any suitable data storage technology, such as semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory. Although only one memory 102 is shown for computer 100, computer 100 may have several physically different memory modules. Processor 101 may be of any type. Processor 101 may include one or more general-purpose computers, dedicated computers, microprocessors, digital signal processors, and, in non-limited examples, processors based on multicore processor architectures. Computer 100 may have multiple processors, such as application-specific integrated circuit chips that are time-dependent to a clock that synchronizes the main processor.

[0048] Embodiments of this disclosure may be implemented in hardware or dedicated circuitry, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor or other computing device.

[0049] This disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, that are executed on a device on a target actual processor or virtual processor to execute the processes or methods of this disclosure. Program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The functions of program modules may be combined or divided among program modules as desired in various embodiments. The machine-executable instructions of program modules can be executed within a local or distributed device. In a distributed device, program modules can be arranged on both local and remote storage media.

[0050] The program code for executing the methods of this disclosure may be written in any combination of one or more programming languages. These program codes are provided to a processor or controller of a general-purpose computer, a dedicated computer, or other programmable data processing devices. When the program code is executed by the processor or controller, the functions / operations in the flowchart and / or the implementation block diagram are executed. The program code is executed entirely on the machine, partially on the machine as a stand-alone software package, partially on the machine, partially on a remote machine, or entirely on a remote machine or server.

[0051] Programs can be stored and supplied to a computer using various types of non-temporary computer-readable media. Non-temporary computer-readable media include various types of tangible recording media. Examples of non-temporary computer-readable media include magnetic recording media, magneto-optical recording media, optical disc media, and semiconductor memory. Magnetic recording media include, for example, flexible disks, magnetic tapes, and hard disk drives. Magneto-optical recording media include, for example, magneto-optical disks. Optical disc media include, for example, Blu-ray discs, CD (Compact Disc)-ROM (Read Only Memory), CD-R (Recordable), and CD-RW (ReWritable). Semiconductor memory includes, for example, solid-state drives, mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory). Programs may also be supplied to a computer using various types of temporary computer-readable media. Examples of temporary computer-readable media include electrical signals, optical signals, and electromagnetic waves. Temporary computer-readable media can supply programs to a computer via wired communication channels such as electric wires and optical fibers, or via wireless communication channels.

[0052] [Description of Effects] The information processing device 10 in Embodiment 2 achieves the same effects as the information processing device 10 in Embodiment 1. Furthermore, the information processing device 10 in Embodiment 2 can efficiently collect and analyze large amounts of information. The acquisition unit 110 collects information from various sources, and the analysis unit 120 analyzes it, enabling analysis of a scale and depth that would be difficult to achieve manually. This allows for a more accurate understanding of market trends and consumer needs.

[0053] (Embodiment 3) Embodiment 3 will now be described. Embodiment 3 differs from the configurations of Embodiments 1 and 2 in that the analysis unit further comprises a specific unit.

[0054] [Explanation of the premise] In new product development, it is important to accurately grasp market trends and consumer needs and identify products and services that deserve attention. However, obtaining important insights from vast amounts of market data and consumer information has been a time-consuming and laborious challenge. Therefore, this embodiment aims to support developers' decision-making by automatically identifying products and services that deserve attention from the results of data analysis.

[0055] [Description of Configuration] The outline of the information processing device 20 will be described with reference to Figure 4. Figure 4 is a block diagram showing an example of the information processing device 20 in this embodiment. The information processing device 20 comprises an acquisition unit 210, an analysis unit 220, and a display unit 230. The analysis unit 220 comprises a identification unit 221. Since the basic configuration of the acquisition unit 210 and the display unit 230 is the same as in Embodiments 1 and 2, the analysis unit 220 will be described in detail here.

[0056] The acquisition unit 210 and the display unit 230 have the same functions as the acquisition unit 110 and the display unit 130 in Embodiments 1 and 2.

[0057] The analysis unit 220 performs analysis based on the information collected by the acquisition unit 210. The analysis unit 220 can have the same functions as the analysis unit 120 in Embodiments 1 and 2. The analysis unit 220 also includes a identification unit 221. The identification unit 221 identifies products or services that are noteworthy in the market based on the results of the analysis performed by the analysis unit 120 as described in Embodiments 1 and 2.

[0058] The identification unit 221 identifies products or services that are noteworthy in the market (products or services that meet the attention criteria) based on the analysis results obtained by the analysis unit 220. The identification methods can vary, and the identification unit 221 may use at least one of the following: sales analysis, sentiment analysis, trend analysis, and competitive analysis. The identification unit 221 may also use multiple analysis methods for identification. The attention criteria may be predetermined.

[0059] In sales analysis, the identification unit 221 uses data acquired by the acquisition unit 210 to identify products or services with rapidly increasing sales. The identification unit 221 may also detect products or services with rapidly increasing sales as those that deviate from the normal growth curve by a predetermined level or more in the direction of increasing sales. Furthermore, the identification unit 221 may track changes in growth rate and market share to identify rapidly growing products or services and products or services that continue to grow steadily. The following are examples of conditions for rapidly growing products or services and products or services that continue to grow steadily, but are not limited to these examples.

[0060] For example, the identification unit 221 may identify a product or service whose growth rate is above a first threshold as a rapidly growing product or service. In addition, the identification unit 221 may identify a product or service whose "growth rate is above the first threshold" has continued for a predetermined period of time or longer as a rapidly growing product or service. Furthermore, the identification unit 221 may identify a product or service whose "growth rate is above a second threshold but below the first threshold" has continued for a predetermined period of time or longer as a product or service that is experiencing stable growth.

[0061] In addition, the specific unit 221 may identify a product or service whose market share growth rate is above the third threshold as a rapidly growing product or service. In addition, the specific unit 221 may identify a product or service whose "market share growth rate is above the third threshold" has continued for a predetermined period of time or longer as a rapidly growing product or service. In addition, the specific unit 221 may identify a product or service whose "market share growth rate is above the fourth threshold but below the third threshold" has continued for a predetermined period of time or longer as a product or service that is experiencing stable growth.

[0062] In sentiment analysis, the identification unit 221 uses natural language processing technology to calculate sentiment scores from consumer reviews of products or services and identifies products or services that have received high ratings. For example, the identification unit 221 may use a language model generated by machine learning to perform highly accurate sentiment analysis that takes context into account. As a result, the identification unit 221 may quantify consumer voices and identify products or services that have received favorable ratings (products or services that evoked positive emotions) or products or services that have room for improvement (products or services that evoked negative emotions).

[0063] Trend analysis involves analyzing search data and social media mentions to identify products or services that are gaining attention. The identification unit 221 may also identify fields, products, or services that are attracting consumer interest based on trends in search keywords and the frequency of hashtag usage. For example, the identification unit 221 may identify words related to fields, products, or services among words whose search word usage ranking or hashtag usage ranking meets predetermined conditions as fields, products, or services that are attracting consumer interest. These predetermined conditions include, but are not limited to, a ranking above a threshold.

[0064] In competitive analysis, the identification unit 221 performs an analysis of the competitors of the company to which the user performing the analysis belongs, using the information processing device 20. At least one company may be registered in the information processing device 10 as a competitor in advance. The identification unit 221 may then identify the competitor based on the registered information. Alternatively, the identification unit 221 may identify the competitor by querying the generating AI.

[0065] The identification unit 221 can identify products that are attracting attention within the industry based on information such as fluctuations in the market share of competing companies and announcements of new products by competing companies. The identification unit 221 may also identify the characteristics of successful products in the market and areas where future growth is expected by analyzing the trends of the competing companies and the overall industry trends based on the data acquired by the acquisition unit 210. For example, the identification unit 221 may input information about the competing companies (such as trends and data indicating the overall industry trends) into the generating AI and have it identify the characteristics of successful products in the market and areas where future growth is expected based on that data.

[0066] The identification unit 221 may use at least two of these analytical methods in combination to identify multiple products or services as candidates for new product development. For example, the identification unit 221 may weight the results of each analytical method and calculate an integrated score. Specifically, it may weight sales analysis at 40%, sentiment analysis at 30%, trend analysis at 20%, and competitor analysis at 10% to identify products or services with a high overall score. Alternatively, the identification unit 221 may first extract products with high growth rates using sales analysis, and then filter those products using sentiment analysis. Furthermore, the identification unit 221 may further narrow down the candidates in stages, such as by focusing on products showing increased attention using trend analysis. By combining multiple analytical methods, promising candidates for new product development can be identified. In addition, these methods can be adjusted as appropriate according to market changes and corporate strategies.

[0067] The identification unit 221 may analyze at least one of the success factors and failure factors for the identified product or service. Analysis methods may include feature extraction, correlation analysis, analysis of A / B test results, and analysis of failure cases.

[0068] In feature extraction, the identification unit 221 extracts features such as product functions, design, and price range, and identifies commonalities among successful products. For example, the identification unit 221 can use text mining technology to extract key functions from product descriptions, or use image recognition technology to quantify design features. The identification unit 221 may also analyze features related to market strategies, such as price range and sales channels, as targets for extraction.

[0069] In correlation analysis, the identification unit 221 analyzes the correlation between each feature (extracted commonalities) and sales or customer satisfaction to identify which elements contribute to success. For example, the identification unit 221 can quantitatively evaluate the extent to which a particular function or price range affects sales using Pearson's correlation coefficient or regression analysis. Alternatively, the identification unit 221 may use multivariate analysis to analyze the impact when multiple factors are combined.

[0070] In analyzing A / B test results, the specific unit 221 analyzes data from past A / B tests to identify which changes improved customer response. For example, the specific unit 221 can statistically analyze the impact of small changes, such as website design changes or package color changes, to identify effective areas for improvement. The specific unit 221 may also integrate the results of multiple A / B tests to analyze trends on a larger scale.

[0071] In analyzing failure cases, the specific unit 221 analyzes the characteristics (commonalities, etc.) of products that have been withdrawn from the market or that have not achieved the expected results, and identifies factors to be avoided. For example, the specific unit 221 can predict and avoid future risks by creating a database of past failure cases and extracting failure patterns using a machine learning algorithm. Information on past failure cases can include information such as the market environment at the time, product characteristics, and marketing strategy. The specific unit 221 may also analyze the causes of failure from multiple perspectives, such as the market environment, product characteristics, and marketing strategy. For example, the specific unit 221 may input the above failure case information into a generating AI and have it analyze the causes of failure from multiple perspectives, such as the market environment, product characteristics, and marketing strategy, based on that information.

[0072] [Explanation of Operation] Figure 5 is a flowchart showing an example of the operation of the information processing device 20. An example of a series of processes performed by the information processing device 20 will be explained with reference to Figure 5.

[0073] The acquisition unit 210 collects information related to product development (step S210). In this process, the acquisition unit 210 collects various types of data, such as market data, customer reviews, sales performance, competitor information, and other information mentioned above.

[0074] The analysis unit 220 performs an analysis of the collected information on product development (step S220). In this analysis process, the analysis unit 220 extracts market trends, customer preferences, sales patterns, etc., using methods such as data mining, statistical analysis, and text analysis.

[0075] The identification unit 221 identifies products and services of interest based on the analysis results of step S220 (step S230). In this process, the identification unit 221 uses methods such as sales analysis, sentiment analysis, trend analysis, and competitor analysis to identify products that are successful or rapidly growing in the market. The identification unit 221 may also analyze the success factors and failure factors of the identified products. Note that the order of steps S220 and S230 is not limited; identification may be performed before analysis, or analysis may be performed before identification.

[0076] The display unit 240 displays the analysis results from step S220 and the product or service identified in step S230 (step S240). The display unit 240 may further display the success factors and failure factors analyzed in step S230. The display method is the same as in embodiments 1 and 2, but more detailed information may be clearly shown by also displaying the analysis results of the identified product of interest and success factors.

[0077] [Description of Effects] The information processing device 20 in Embodiment 3 achieves the same effects as the information processing device 10 in Embodiments 1 and 2. Furthermore, the information processing device 20 in Embodiment 3 automatically collects and analyzes large amounts of market data and customer information, providing objective insights that enable product development that better suits market needs. In addition, by extracting important information and presenting it as identified products or factors, the development team can concentrate its limited resources on promising directions.

[0078] (Embodiment 4) Embodiment 4 will now be described. In this embodiment, information on the company's own product group and the competitor's product group, as well as market information, is used as input to determine the success or failure of a product, and a process is described to analyze the factors contributing to the success of competitors' products and the differences between similar products of the company. Note that explanations that overlap with Embodiments 1 to 3 will be omitted.

[0079] [Explanation of the premise] In new product development, it is extremely important to compare and analyze one's own products with those of competitors and to accurately understand their position in the market. However, it has been difficult to integrate and analyze diverse product information and market data to gain meaningful insights. Therefore, this embodiment aims to support the formulation of more effective product development strategies by integrating information on one's own product group, competitor product group, and market information to clarify the success and failure factors of a product.

[0080] [Description of Configuration] Figure 6 is a block diagram showing the configuration of the information processing device 30. The information processing device 30 comprises an acquisition unit 310, a feature quantity generation unit 320, an analysis unit 330, and a display unit 340. The analysis unit 330 comprises a specific unit 331.

[0081] The acquisition unit 310 acquires information on the company's own product group and the product group of competitors (POS data, customer feedback, etc.) and market information. Specifically, the acquisition unit 310 may acquire information such as sales figures, sales volume, and price for the company's own products and the competitors' products from POS data. The acquisition unit 310 may also acquire text data such as product evaluations and comments from customer feedback. Furthermore, the acquisition unit 310 may acquire industry reports and market research data as market information. The acquisition unit 310 may also have the same functions as the acquisition unit 110 in Embodiments 1 and 2 and the acquisition unit 210 in Embodiment 3.

[0082] The feature generation unit 320 generates features that may influence the prediction target (e.g., sales) from the data acquired by the acquisition unit 310. The feature generation unit 320 generates a feature generation function that generates features using a first table (e.g., product information table) and a second table (e.g., POS data table). This feature generation function is generated by applying various parameters to a pre-prepared template. The generated features may include, for example, the average selling price of a product, monthly sales, and the average customer rating.

[0083] This section explains the definition of market information and provides an example of a method for generating features using it. Market information refers to all data related to the sales environment of a product or service. For example, market information includes economic indicators, demographic data, competitive information, consumer trend data, and external environmental factors. Economic indicators include, for example, the Consumer Price Index, GDP (Gross Domestic Product), disposable income, and exchange rates. Demographic data includes population by age group, household composition, income distribution, and characteristics of residential areas. Competitive information includes competitors' product prices, new product launch dates, sales promotion activities, and market share. Consumer trend data includes purchasing behavior data, product reviews, mentions on social media, and search trends. External environmental factors include weather data, seasonal factors, event information, and regulatory changes.

[0084] To utilize this market information, the feature generation unit 320 can use a predetermined source code as a database extraction template. An example of the source code is shown below, but is not limited to this. "SELECT Product ID, Date, AVG(Sales Quantity) as Average Sales, SUM(Sales Amount) as Total Sales, COUNT(DISTINCT Buyer ID) as Number of Buyers FROM Sales Data JOIN Market Information ON Sales Data.Date = Market Information.Date WHERE Specified Analysis Period GROUP BY Specified Aggregation Unit".

[0085] The feature generation unit 320 uses the above template as a query and sets parameters such as the analysis period and aggregation unit. For example, the analysis period is set to "First Quarter of 2024" and the aggregation unit to "Weekly". These parameters may be set in advance or can be entered by the user.

[0086] The feature generation unit 320 calculates features from the acquired data, such as a "moving average of the past three months." At this time, the generated features may be normalized by using a relative index with the market average set to 100 or by standardizing them to a range of 0 to 1.

[0087] The analysis unit 330 uses the generated feature quantities to perform a factor analysis of the success or failure of a product. In this embodiment, the analysis unit 330 uses XAI as the analysis method, but other methods may be used for analysis. Specifically, the analysis unit 330 analyzes the factors for the success or failure of a product based on the generated feature quantities, using sales, customer satisfaction, market share, etc. as criteria. Furthermore, the analysis unit 330 can use XAI to perform a comparative analysis of the company's own products and competitors' products to clarify the differences between them. The analysis unit 330 may have the same functions as the analysis unit 120 in Embodiments 1 and 2 and the analysis unit 220 in Embodiment 3.

[0088] The identification unit 331 identifies products and services that are noteworthy in the market based on the analysis results of the analysis unit 330. The identification unit 331 can achieve this identification using the same processing as the identification unit 221 in Embodiment 3. Furthermore, the identification unit 331 may have the same functions as the identification unit 221 in Embodiment 3.

[0089] The display unit 340 visualizes and displays the analysis results, identification results, and difference analysis results. The display unit 340 may have the same functions as the display unit 130 in Embodiments 1 and 2 and the display unit 230 in Embodiment 3.

[0090] [Explanation of Operation] Figure 7 is a flowchart showing an example of the operation of the information processing device 30. An example of a series of processes performed by the information processing device 30 will be explained with reference to Figure 7.

[0091] First, the acquisition unit 310 acquires information on the company's own product group and the product group of other companies, as well as market information (step S310).

[0092] Next, the feature generation unit 320 generates features (step S320). First, it generates a feature generation function by applying table information such as the POS data table name, join information by product ID (identifier), aggregation information such as the total sales, and selection information such as data selection for a specific period to a pre-prepared first template. Join information refers to information that defines the relationships between different tables in the database. For example, by joining the product master table and the POS data table using the product ID, detailed product information and sales performance data can be combined and analyzed. When the feature generation unit 320 generates features for multiple tables, it can refer to this join information to understand the relationships between the tables.

[0093] The first template provides a framework for generating a feature generation function. This template may include the basic structure of an SQL statement. For example, the first template may consist of some or all of the table specification section, join condition section, aggregate function section, grouping section, and period specification section. Next, the feature generation unit 320 uses this function to generate features from the product information table and the POS data table. Furthermore, the feature generation unit 320 uses the second template to create a document explaining the meaning of the generated features. The second template may provide a framework for creating a document explaining the meaning of the generated features. This template may be written in natural language. For example, the second template may consist of some or all of the feature name section, calculation method explanation section, target period section, and semantic interpretation section. The specific structure and content of the templates may be modified as appropriate depending on the analysis target and purpose, and are not limited to the above examples.

[0094] Next, the analysis unit 330 uses XAI to perform a factor analysis of the success or failure of a product, taking the features generated by the feature generation unit 320 as input (step S330). The analysis unit 330 also uses XAI to perform a comparative analysis of the company's own products and those of competitors.

[0095] For example, the analysis unit 330 takes features as input and constructs a machine learning model that outputs product success indicators (e.g., sales revenue, market share). This model is trained using data from both the company's own products and those of competitors. Next, the analysis unit 330 applies XAI to the constructed model to analyze, in an interpretable way, how much each feature contributes to the success or failure of the product. This makes it possible to clearly visualize the differences in success factors between the company's own products and those of competitors.

[0096] The identification unit 331 identifies products and services that deserve attention in the market based on the analysis results (step S340).

[0097] Finally, the display unit 340 visualizes and displays the analysis results and identification results (step S350). For example, the display unit 340 may display a graph comparing the feature distributions of successful and unsuccessful products, a ranking of the importance of features, a radar chart comparing the features of successful products from other companies with similar products from the same company, etc. The display unit 340 may also display a list of noteworthy products or services and their characteristics. Furthermore, the display unit 340 may add generated explanatory text to each feature to make it easier for the user to understand the meaning of the features.

[0098] [Description of Effects] The information processing device 30 in Embodiment 4 achieves the same effects as the information processing devices 10 and 20 in Embodiments 1 to 3. Furthermore, the information processing device 30 in Embodiment 4 can automate detailed comparative analysis of the company's own products and competitors' products, and objectively identify success factors and areas for improvement. In addition, it becomes possible to identify noteworthy products and services in the market from multiple perspectives, supporting more strategic product development and market entry decisions.

[0099] (Embodiment 5) Embodiment 5 will now be described. In this embodiment, a hypothesis generation unit is added to the configuration of Embodiments 1 to 4, and the process of generating hypotheses for product development based on the analysis results will be described.

[0100] [Explanation of the premise] In new product development, while there is information that can be visualized using conventional methods such as surveys and interviews, it is important to capture consumers' latent needs and desires. Therefore, in this embodiment, a hypothesis generation unit is introduced. The purpose of the hypothesis generation unit is to derive consumer insights from the analysis results. Here, insights refer to the deep-seated motivations and values ​​behind consumers' behavior and thinking. This means understanding consumers' essential needs and desires, not just aggregating data or grasping superficial trends.

[0101] [Description of Configuration] Figure 8 is a block diagram showing the configuration of the information processing device 40. This embodiment is also applicable to embodiments 1 to 3, but here we will describe an embodiment in which a hypothesis generation unit is added to embodiment 4. The information processing device 40 comprises an acquisition unit 310, a feature quantity generation unit 320, an analysis unit 330, a hypothesis generation unit 440, and a display unit 340. The analysis unit 330 comprises a specification unit 331. Since the functions of the acquisition unit 310, feature quantity generation unit 320, analysis unit 330, specification unit 331, and display unit 340 are the same as in the embodiments described above, the hypothesis generation unit 440 will be described in detail here.

[0102] The hypothesis generation unit 440 generates hypotheses for new product development based on the analysis results of the analysis unit 330. The hypothesis generation unit 440 may also generate hypotheses for new products based on market needs and consumer trends identified from the analysis results. Alternatively, the hypothesis generation unit 440 may generate hypotheses for new products based on combinations of success factors that generate hypotheses for innovative new products by combining the characteristics of existing successful products in a new way.

[0103] Furthermore, the hypothesis generation unit 440 may combine the latest technological trends with existing product categories to generate hypotheses for products that offer new value. In addition, the hypothesis generation unit 440 may identify market segments not covered by the current product lineup and generate hypotheses for new products for untapped markets.

[0104] The hypothesis generation unit 440 may generate hypotheses for new products by combining characteristics of different product categories. Alternatively, the hypothesis generation unit 440 may analyze consumers' potential needs and challenges and generate hypotheses for new products based on consumer insights.

[0105] The hypotheses for new products generated by the hypothesis generation unit 440 may include ideas for new products that are expected to be successful in the market. These hypotheses may predict at least one of the new product's concept and features. These hypotheses may include a product concept that shows the basic idea and purpose of the new product, a target customer base that is an expected main customer segment, and key features and characteristics that represent the main value and functions that the product offers. Furthermore, they may include differentiating factors that show the differences and advantages from competing products, expected pricing, sales channels that indicate effective sales methods and locations, a marketing strategy that indicates the direction of promotion and branding, and potential challenges such as expected development and sales challenges.

[0106] The hypothesis generation unit 440 may utilize a large-scale language model to process the analysis results of the analysis unit 330 and generate a detailed hypothesis that includes the elements described above. For example, the hypothesis generation unit 440 can generate a hypothesis such as the following: "A multi-functional cosmetic product targeting working women in their 20s and 30s living in urban areas, combining portability and functionality. It combines the three functions of conventional cosmetics into one, with a compact design to simplify touch-ups during the day. It uses environmentally friendly, renewable materials and is compliant with the SDGs. The price range is mid-to-high, and sales will mainly be through specialty stores and online."

[0107] [Description of Operation] Figure 9 is a flowchart showing an example of the operation of the information processing device 40. Since steps S410 to S440 are the same as in the embodiment described above, the operation step S450 of the hypothesis generation unit 440 will be described here.

[0108] The hypothesis generation unit 440 first receives the analysis results from the analysis unit 330 as input. Depending on the embodiment, these analysis results include various information such as the success factors of the product, market trends, and consumer needs. Next, the hypothesis generation unit 440 generates a hypothesis for the new product based on this information (step S450).

[0109] The hypothesis generation unit 440 may output insights based on the sales factors and acquired information identified by the analysis unit 330 using generating AI. For example, suppose the analysis unit 330 analyzes the sales of a particular food product and identifies "women in their 20s" and "living in the city center" as factors contributing to its sales. The hypothesis generation unit 440 performs an analysis using the social media data acquired by the acquisition unit 310 and the keywords "women in their 20s" and "living in the city center." As a result, the hypothesis generation unit 440 can identify, for example, that the food product was frequently advertised on social media where women in their 20s are the main users at a certain time. Based on this identification result, the hypothesis generation unit 440 can then generate insights such as the following using generating AI: "It is thought that the strategic advertising of this product on social media where women in their 20s are the main users contributed to the increase in sales. It is presumed that it appealed to the needs of a specific group of women who want to live a sophisticated life in the city. It is thought to reflect the consumer insights of this group who seek a fashionable lifestyle and a sophisticated image."

[0110] In this way, the hypothesis generation unit 440 can generate hypotheses that link data-based analysis results with consumers' underlying psychology. The multiple hypotheses generated by the hypothesis generation unit 440 may be evaluated and prioritized from the perspectives of marketability, technical feasibility, and consistency with corporate strategy. The hypothesis generation unit 440 may then list the high-priority hypotheses and add comments indicating the rationale and expected market impact for each hypothesis.

[0111] The display unit 340 visualizes and displays the analysis results and the generated new product development hypotheses (step S460).

[0112] [Description of Effects] The information processing device 40 in Embodiment 5 achieves the same effects as the information processing devices 10 to 30 in Embodiments 1 to 4. Furthermore, the information processing device 40 in Embodiment 5 can generate product ideas using the hypothesis generation unit 440. By generating hypotheses from multiple perspectives, it becomes possible to indicate more comprehensive and diverse directions for new product development. Regardless of whether it is applied to Embodiments 1, 2, or 3, this embodiment is common in that it creates new value based on existing analysis results, and can strengthen the new product development process.

[0113] (Embodiment 6) Embodiment 6 will now be described. In this embodiment, the input information and output information of the information processing device 30 of Embodiment 4 will be described further.

[0114] [Explanation of the premise] In new product development, analyzing the visual elements of a product and its market exposure is important for understanding consumer purchasing behavior. However, it has been difficult to fully grasp the actual display situation of a product and its interaction with consumers using only text data and numerical data. Therefore, this embodiment aims to support the formulation of an effective product development strategy by obtaining and analyzing product information from various sources, including image data.

[0115] [Description of Configuration] The basic configuration of the information processing device 30 in this embodiment is the same as that of the information processing device 30 in Embodiment 4. Therefore, redundant explanations will be omitted.

[0116] The acquisition unit 310 can acquire image data in addition to text data and numerical data as information about a product. The acquisition unit 310 may also have a function to identify a product from the acquired image data.

[0117] The acquisition unit 310 identifies products using image recognition technology. Specifically, the acquisition unit 310 may identify products by object detection using a convolutional neural network (CNN), product classification using a deep learning model, or extraction of visual features. In addition, the acquisition unit 310 may further identify products using technologies such as text information extraction using OCR (Optical Character Recognition).

[0118] This process allows the acquisition unit 310 to identify products from screenshots of store shelves, online shops, and even images on social media, and to utilize additional information such as their placement and visibility for analysis.

[0119] Furthermore, the acquisition unit 310 may extract the shooting date and time (time the image was taken) and location information from the EXIF ​​(Exchangeable Image File Format) data of each of the multiple images containing the product, thereby improving the accuracy of product identification. For example, if the image was taken at a specific store, the acquisition unit 310 can more accurately identify the product by comparing it with the store's inventory information. In addition, the acquisition unit 310 can analyze other products and background information within the image to extract contextual information such as the product's placement and display method. This can be used to analyze the product's market positioning and competitive situation.

[0120] This section describes a specific example of a method to improve the accuracy of product identification from the date and time or location information of an image. The acquisition unit 310 first extracts EXIF ​​data from the image file. The EXIF ​​data includes information such as the camera settings at the time of shooting, as well as the date and time of shooting (DateTime), GPS coordinates (GPSLatitude, GPSLongitude), and shooting direction (GPSImgDirection).

[0121] For example, the acquisition unit 310 converts GPS coordinates into geographical address information (prefecture, city, ward, town, or village) using reverse geocoding. Reverse geocoding is a technique that identifies a corresponding address from latitude and longitude coordinates. For example, the acquisition unit 310 can obtain the address information "Marunouchi, Chiyoda-ku, Tokyo" from coordinate data of latitude 35.6812362 and longitude 139.7671248. The acquisition unit 310 uses this location information to identify the store where the image was taken. The acquisition unit 310 compares the GPS coordinates with the store database, and if the coordinates are within the range of a store, it can identify the image as having been taken at that store. In this way, the acquisition unit 310 can identify the store that sells a specific product.

[0122] Furthermore, the acquisition unit 310 may use the shooting date and time information to compare with the product handling history at that store. Specifically, the acquisition unit 310 compares with the store's product master at the time the image was taken and identifies products that match the image recognition result. More specifically, the acquisition unit 310 compares with the store's product master at the time the image was taken and identifies products that match the image recognition result. Even for products with similar packaging or designs, considering the temporal and spatial context makes it possible to identify products with higher accuracy.

[0123] Furthermore, the acquisition unit 310 may be used to analyze the display position of products within a store using information on the shooting direction. For example, even with images of the same product, the visibility and prominence will differ depending on whether the image is taken from the front or from an angle. The acquisition unit 310 can use this information on the shooting direction to more accurately analyze the visual exposure of the product. It becomes possible to quantitatively evaluate whether the product display is in a position that is easily visible from the front, or whether it is placed in a location with high visibility from the aisle.

[0124] In this way, the acquisition unit 310 can accurately identify the company's own products and those of competitors from screenshots of store shelves and online shops, and utilize additional information such as placement and display quantity for analysis. Furthermore, by applying this technology, it is possible to identify products from advertising images and images posted on social media by consumers, and analyze the market exposure of products and consumer usage.

[0125] The feature generation unit 320 generates features including information extracted from image data. For example, the placement of a product within a store or the frequency of product exposure in social media posts (cumulative exposure count, exposure count within a predetermined period, etc.) may also be treated as features.

[0126] The analysis unit 330 may perform a more comprehensive factor analysis by adding information obtained from image data. For example, the analysis unit 330 may analyze the correlation between product display location and sales, or the relationship between social media exposure and market share.

[0127] The identification unit 331 identifies noteworthy products and services that meet predetermined conditions, taking into consideration the results of the image data analysis. For example, the identification unit 331 may identify products that show a correlation between prominent display locations in stores and high sales as the focus of analysis.

[0128] The display unit 340 displays analysis results and product information in various formats and also has a function to search and display information based on conditions specified by the user.

[0129] [Description of Operation] Figure 10 is a flowchart showing an example of the operation of the information processing device 30. Note that the same processes as in Embodiment 4 will not be described, and the explanation will focus on the parts that have been extended in this embodiment.

[0130] The acquisition unit 310 acquires information on the company's own product group and the product group of other companies, as well as market information (step S510). In this step, the acquisition unit 310 acquires image data in addition to the data acquired in Embodiment 4. For example, the acquisition unit 310 accepts as input the like: photographs of store shelves, screenshots of e-commerce sites, product images posted on social media, etc.

[0131] Next, the acquisition unit 310 identifies the product from the acquired image data (step S520). The acquisition unit 310 may also extract additional information such as the product's placement and display quantity. For example, the acquisition unit 310 may identify the location and quantity of each product from images of store shelves, or analyze product usage from images posted on social media.

[0132] Next, the feature generation unit 320 generates features (step S530). This step is the same as in Embodiment 4, but differs in that it also generates features including information extracted from the image. For example, the feature generation unit 320 may also treat the placement of the product within the store or the frequency of the product's exposure in social media posts as features.

[0133] The analysis unit 330 uses XAI to analyze the factors contributing to the success or failure of a product, taking the features generated by the feature generation unit 320 as input (step S540). Specifically, the analysis unit 330 can analyze a product based on criteria such as sales revenue, customer satisfaction, and market share. Furthermore, the analysis unit 330 can perform a comparative analysis of the company's own products with those of competitors. In this analysis, the analysis unit 330 may also consider information obtained from image data (such as the correlation between product display location and sales, and the relationship between social media exposure and market share).

[0134] Based on the analysis results, the identification unit 331 identifies products and services that are noteworthy in the market (step S550). For example, the identification unit 331 may identify products that show high sales (products with sales above a threshold, etc.). In addition, the identification unit 331 may identify products that are seeing a surge in mentions on social media (products with an increase speed above a threshold, etc.). In addition, the identification unit 331 may identify products that show a correlation between prominent display locations in stores and high sales (such as products where the correlation is above a standard level). Note that the products identified by the identification unit 331 are not limited to the examples given here.

[0135] Finally, the display unit 340 visualizes and displays the analysis results, identification results, and difference analysis results (step S560).

[0136] For example, the display unit 340 may display product information linked to the image recognition results. Specifically, the display unit 340 may overlay product information onto an image of a store shelf taken by a user and inputted into the information processing device 30. For example, when a user inputs an image of a store shelf into the information processing device 30, the information processing device 30 can analyze the image and recognize the products. The display unit 340 can then overlay product information for each product, linked to each product, onto the image. For example, when the display unit 340 receives user input by tapping the icon of each recognized product, it may display detailed information (product information) such as sales figures, market share, and customer ratings for that product in a pop-up window. Furthermore, the display unit 340 may analyze the product placement and display quantity and display a suggestion for the optimal product layout.

[0137] Furthermore, the display unit 340 may visually display the results of the social media analysis. For example, the display unit 340 may represent the sentiment analysis results of social media posts about a specific product as a pie chart showing the proportion of positive, negative, and neutral sentiments. Alternatively, the display unit 340 may show the trend in the amount of mentions on social media over a predetermined period (e.g., the past six months) as a line graph and display the correlation with a specific event or advertising campaign. In addition, the display unit 340 may display frequently occurring keywords related to the product as a word cloud, allowing consumers to visually understand their interests and the product's distinctive evaluation points.

[0138] For example, the display unit 340 may display a list of multifaceted information about each product. In this example, the display unit 340 can place a product list on the left side of the screen and display detailed information (product information) of the product selected by user input on the right side of the screen. The detailed information may include graphs of monthly sales data, market share trends, customer satisfaction scores, mentions on major social media platforms, and feature comparison tables with competing products. The display unit 340 may also display sales forecasts and market trend forecasts based on machine learning, providing information useful for future strategic planning.

[0139] The display unit 340 may also provide a factor analysis tool. For example, the display unit 340 may place sliders at the top of the screen for key factors influencing the success of a product (such as price, quality, marketing investment, and distribution channels). The display unit 340 may then display graphs of the product's success probability and projected sales at the bottom of the screen. When the display unit 340 receives user input to adjust the sliders, it can change the content of the graphs of the product's success probability and projected sales in real time. The display unit 340 can also consider the interactions of each factor and visualize the impact on other factors when one factor is changed. Furthermore, the display unit 340 may provide a comparison simulation function with competing products and display a prediction of changes in market share when each factor of the company's own product is adjusted.

[0140] [Description of Effects] The information processing device 30 in Embodiment 6 achieves the same effects as the information processing devices 10 to 40 in Embodiments 1 to 5. Furthermore, this embodiment allows users to gain an overview of the entire product lifecycle while also accessing detailed information at each stage, thus enabling comprehensive information gathering. In particular, the use of image data makes it possible to include information that was previously difficult to capture, such as product placement in stores and actual consumer usage, in the analysis. This is expected to lead to more realistic strategic planning and the discovery of potential needs in new product development.

[0141] (Embodiment 7) Embodiment 7 will now be described. This embodiment differs from Embodiments 1 to 6 in that it further includes a design generation unit in addition to the configuration of any of Embodiments 1 to 6. In this embodiment, the process of utilizing the analysis results obtained in the previous embodiments for product design generation will be described.

[0142] [Explanation of the premise] In new product development, effectively reflecting market analysis results in product design is crucial for creating products that meet consumer needs. However, the process of deriving design concepts from analysis results and translating them into concrete visual representations heavily relies on the designer's skills and subjectivity, sometimes resulting in discrepancies with the analysis results. Therefore, this embodiment aims to automatically generate product design proposals based on analysis results and realize a data-driven design development process.

[0143] [Description of Configuration] Figure 11 is an example of a block diagram showing the configuration of the information processing device 60. The information processing device 60 of this embodiment differs from the information processing devices 10 to 40 of embodiments 1 to 6 in that it includes a design generation unit 610. Except for the inclusion of the design generation unit 610, the information processing device 60 can have the same configuration as the information processing devices 10 to 40 of embodiments 1 to 6. The design generation unit 610 will be described in detail below.

[0144] The design generation unit 610 receives the various analysis results described in Embodiments 1 to 6 as input and utilizes these results to generate the product design. Specifically, the design generation unit 610 consists of a text generation unit 611 and a label generation unit 612.

[0145] The text generation unit 611 generates text describing the analysis results and the characteristics of identified products using a machine learning model. This machine learning model combines a data analysis model and a natural language processing model. For example, the results of factor analysis by the analysis unit 330, the results of identifying noteworthy products by the identification unit 331, and hypotheses for new product development by the hypothesis generation unit 440 become inputs to this machine learning model. Based on these inputs, the machine learning model generates text describing the characteristics of the new label design. For example, this machine learning model generates and outputs text describing the characteristics of the new label design, such as, "A light pink package that gives a sophisticated impression, aimed at women in their 20s. It features a geometric design that symbolizes an urban lifestyle."

[0146] The label generation unit 612 uses an image generation model (generation AI) with the text generated by the text generation unit 611 as input to generate a new product image. The image generation model generates a product label image that reflects the characteristics of the input text while adding new design elements. The label generation unit 612 is not limited to generating product labels, but can also generate various visual elements related to the product, such as package design, product body design, user interface design, and advertising visuals. For example, the label generation unit 612 may generate a 3D model of the product or propose a website layout design based on the text input.

[0147] [Explanation of Operation] Figure 12 is a flowchart showing an example of the operation of the information processing device 60.

[0148] First, the information processing device 60 performs processing such as data acquisition, feature generation, factor analysis, identification of products or services of interest, and hypothesis generation for new product development, as described in any of embodiments 1 to 6, and outputs the analysis results (step S610).

[0149] Next, the text generation unit 611 of the design generation unit 610 receives the outputted analysis results as input (step S620).

[0150] The text generation unit 611 generates text describing the product's features based on the received analysis results (step S630). In this process, the text generation unit 611 expresses the important features extracted from the analysis results (e.g., target audience, success factors, differentiation points, etc.) in natural language. For example, the text generation unit 611 generates text such as, "A sophisticated product aimed at women in their 20s. It features a geometric design that symbolizes an urban lifestyle and is 30% lighter than competing products."

[0151] Next, the label generation unit 612 receives the generated text as input (step S640).

[0152] The label generation unit 612 generates an image of a new product label using an image generation model based on the received text (step S650). In this process, design elements (color, shape, layout, etc.) that visually represent the features described in the text are automatically selected and combined.

[0153] The generated product label image is sent to the display unit 340. The display unit 340 displays the received product label image (step S660). At this time, the display unit 340 may also display the original analysis results and generated text to clearly show the basis for the design.

[0154] Furthermore, the text generation unit 611 may be provided with a function to receive user feedback. When a user inputs an evaluation or a request for modification regarding the displayed design, that information is sent to the text generation unit 611, and the text may be modified (returning to step S630). By repeating this process, it becomes possible to generate designs that better meet the user's needs.

[0155] [Examples] Several examples of applications of this embodiment are shown below. These examples specifically demonstrate how the information processing device 60 of Embodiment 7 can effectively reflect various analysis results and consumer insights in design and support strategic and creative product development. Each embodiment explains what kind of analysis results and analysis process information are input and what kind of output is generated based on them.

[0156] [Example 1: Integration of Multiple Analysis Results] Example 1 describes the process of integrating analysis results from multiple market segments to generate a design.

[0157] The input data will be the results of analyses of products for women in their 20s and women in their 30s. The analysis results for products for women in their 20s include that the main target group is women aged 20-25 living in urban areas, that popular colors are pastel pink and light blue, and that social media appeal and compact design are important. The analysis results for products for women in their 30s include that the main target group is women aged 30-35 living in suburban areas and raising children, that popular colors are earth tones and beige, and that functionality and durability are important.

[0158] The text generation unit 611 generates the following text based on these inputs: "A multi-functional product for women aged 25-35. A design that appeals to both urban and suburban areas is required. Color preferences vary by age group, with pastel colors favored by those in their 20s and earth tones by those in their 30s. A design that balances social media appeal with practicality is effective. The multi-functionality of the product should be visually represented while also emphasizing its compactness."

[0159] The label generation unit 612 generates an image of a product label based on this text, including the following design elements: • The basic design is simple and modern, with a gradient color scheme to represent a transition from pastel pink to beige. • The layout features icons indicating the product's multi-functionality arranged in a circle to emphasize its compactness. • The material finish combines a sophisticated sheen with a natural texture.

[0160] [Example 2: Reflecting Time-Series Data] Example 2 describes the process of creating a design that reflects time-series data.

[0161] The input data used will be the results of an analysis of the design characteristics of best-selling products over the past three years and an analysis of consumer trends. The analysis of the design characteristics of best-selling products includes information such as: three years ago, traditional Japanese designs were dominant (60% market share); two years ago, modern and minimalist designs emerged (40% market share); and one year ago, environmentally conscious designs experienced rapid growth (30% market share). The consumer trend analysis includes information such as: environmental awareness has increased by 20% over the past three years; and interest in minimalism has increased by 15% over the past two years.

[0162] The text generation unit 611 generates the following text based on these inputs: "Product design has evolved significantly over the past three years. There has been a shift from traditional Japanese designs to more modern and minimalist designs, and recently, designs emphasizing environmentally conscious materials have emerged. Combining these trends, there is a demand for designs that fuse traditional Japanese beauty, modern minimalism, and environmental considerations. Specifically, this involves effectively arranging the texture of renewable materials such as washi paper, simplified traditional motifs, and symbols indicating recyclability."

[0163] The label generation unit 612 generates an image of a product label based on this text, including the following design elements: • The basic design features a simplified cherry blossom motif on a background with a Japanese paper-like texture. • The color scheme uses a gradient of light green and beige. • The layout places the product name in the center, with a minimal number of environmental symbols surrounding it. • The material has a matte texture reminiscent of recycled paper. • As an additional element, a QR code is placed on part of the package, linking to detailed information about the product's environmental initiatives.

[0164] [Example 3: Visualization of Competitive Analysis Results] Example 3 describes the process of incorporating competitive analysis results into the design.

[0165] The input data will consist of comparative analysis results of our products against those of competitors, and consumer survey results. The comparative analysis results will include information such as our products being 20% ​​superior in functionality, competitors being 10% superior in design, and price competitiveness being equal. The consumer survey results will include information such as 60% prioritizing functionality, 30% prioritizing design, and 10% prioritizing price.

[0166] The text generation unit 611 generates the following text based on these inputs: "A product label that visually emphasizes high functionality while incorporating sophisticated design elements. It effectively places diagrams and icons that show functionality, while simultaneously employing modern and sophisticated color schemes and typography. It also includes elements that demonstrate the reasonableness of the price. It prominently displays the main functions, improves the design, and visually expresses the value commensurate with the price."

[0167] The label generation unit 612 generates an image of a product label based on this text, including the following design elements: • The basic design will have a clean and modern layout. • The color scheme will use a combination of deep blue (suggesting reliability) and silver (expressing advanced technology). • The layout will prominently display icons indicating the product's main functions, each accompanied by a concise explanation. • A simple graph showing a functional comparison with competing products will be included as a graphic element. • A sophisticated sans-serif font will be used for typography to highlight important information. • A "cost-performance" section showing the balance between price and functionality will be included as an additional element.

[0168] [Example 4: Reflecting Consumer Insights] Example 4 describes the process of directly reflecting extracted consumer insights into the design.

[0169] The input data will consist of the results of consumer insights analysis and product category analysis. Consumer insights analysis includes information such as "environmental considerations" being the most important factor, mentioned by 75% of consumers, followed by "health consciousness" being the next most important factor, mentioned by 60% of consumers. Product category analysis includes information such as the demand for organic products in the food and beverage market increasing by 30% year-on-year.

[0170] The text generation unit 611 generates the following text based on these inputs: "A design that strongly appeals to environmental awareness and health consciousness. It emphasizes the recyclable nature of the materials and incorporates natural motifs. It uses green and brown as the base colors, and a pattern based on leaves and tree rings as the background. It clearly indicates that it is an organic product and includes elements that evoke a healthy lifestyle. It visually represents the transparency of the product's raw materials and manufacturing process, thereby gaining consumer trust."

[0171] The label generation unit 612 generates an image of a product label based on this text, including the following design elements: • The basic design will be an organic shape reminiscent of nature. • The color scheme will primarily consist of deep green and warm brown, with a light yellow accent. • The texture will resemble recycled paper or wood. • As graphic elements, leaf and tree ring patterns will be placed in the background. • As icons, simple icons indicating recyclability, organic certification, and health benefits will be included. • As information display, an infographic briefly showing the origin of raw materials and the manufacturing process will be included. • As an additional element, a QR code will allow access to detailed environmental initiatives and health information.

[0172] [Example 5: Personalized Design] Example 5 describes the process of generating a personalized design by combining individual user preference data.

[0173] The input data used will include user profiles, product category analysis, and user behavior analysis results. User profiles include information such as being a male in his 30s, living in an urban area, being a sports enthusiast, and having a high level of interest in technology. Product category analysis includes information such as the increasing demand for high-performance products in the sports drink market. User behavior analysis includes information such as a high frequency of online product information searches and a tendency to share products on social media.

[0174] The text generation unit 611 generates the following text based on these inputs: "A product label for sports-loving men in their 30s, combining high functionality with urban design. It features curves that evoke dynamic movement and a deep blue color reminiscent of a city's nightscape, with icons effectively placed to indicate the product's main functions. It incorporates technological elements, allowing access to detailed product information through a QR code and AR (Augmented Reality) functionality. It also includes visual elements to encourage sharing on social media."

[0175] The label generation unit 612 generates an image of a product label based on this text, including the following design elements: • The basic design will have a streamlined, modern appearance. • The color scheme will be based on a deep blue, with energetic red and yellow accents. • The graphic elements will combine a city skyline with the silhouette of a runner. • Icons indicating key functions (e.g., electrolyte balance, sustained energy supply) will be placed as functional indicators. • An AR marker will be placed as a technological element, and a function will be implemented to display product information in 3D when a smartphone is held over it. • For information access, detailed nutritional information and recommended training menus will be accessible via a QR code. • For social media integration, a link will be included to a function that allows users to easily post photos of the product in use to social media using a dedicated filter.

[0176] [Description of Effects] The information processing device 60 in Embodiment 7 achieves the same effects as the information processing devices 10 to 40 in Embodiments 1 to 6. Furthermore, the information processing device 60 in Embodiment 6 enables design generation that comprehensively utilizes diverse analysis results, realizing data-driven, objective, and creative design decisions. This improves the efficiency and speed of the product development process and facilitates the realization of advanced personalization. Moreover, it enables both brand consistency and innovation, supporting data-driven strategic product development.

[0177] (Modification) Here, a modification applicable to all embodiments is described. This modification shows an example of a screen generated and output by the information processing device 10.

[0178] Figure 13 shows an example of a screen generated and output by the information processing device 10. When the information processing device 10 receives a selection of the "Trending Products" button on the screen shown, it performs an analysis of trending products. For example, based on the acquired product information, the information processing device 10 identifies products or services that are noteworthy in the market. The information processing device 10 then displays the results on the screen. The "Sales Trend" graph on the screen in Figure 13 is the result of the above identification. The sales figures for the identified products are displayed in the graph.

[0179] Furthermore, the screen provides a UI (User Interface) component for performing a product search. The user can input the product name and characteristics into the UI component and perform a search. The information processing device 10 performs a search using the input word as a key and displays the results on the screen (not shown in Figure 13). The target of the search may be information on the internet, information stored in a predetermined database, or product information acquired by the acquisition units 110, 210, and 310.

[0180] Furthermore, the information processing device 10 can perform various analyses based on the acquired product information and display the results on the screen. In the illustrated example, the information processing device 10 performs the latest trend analysis and displays the results on the screen. The information processing device 10 may perform this trend analysis using, for example, generation AI.

[0181] Furthermore, when the information processing device 10 receives a selection of the "Factor Analysis" button on the screen shown in the figure, it performs the "analysis of factors indicating the reasons why the product sold" as described above and displays the results on the screen. Figure 14 shows an example of a screen showing the results of the factor analysis. The information processing device 10 can generate and output a screen like the one shown in Figure 14, for example.

[0182] In the example shown in Figure 14, the information processing device 10 analyzes the factors contributing to increased sales and the main consumer age groups, and displays the results. The factors contributing to increased sales are shown in a pie chart. When the information processing device 10 receives input to select one of the factors displayed in the graph, it displays the detailed analysis results of that factor. In addition, in the example shown in Figure 14, the information processing device 10 also displays on the screen characteristic words extracted from related keywords and consumer feedback.

[0183] Furthermore, when the information processing device 10 receives a selection of the "Hypothesis Generation" button on the screen shown in Figure 13, it generates the aforementioned "Hypothesis for New Product Development" and displays the result on the screen (not shown in Figure 13).

[0184] While this disclosure has been described above with reference to embodiments, it is not limited thereto. The structure and details of this disclosure can be combined with various modifications and embodiments that will be understood by those skilled in the art within the scope of this disclosure.

[0185] Some or all of the above embodiments may also be described as follows, but are not limited to the following: 1. An information processing device comprising: an acquisition means for acquiring information about a product; an analysis means for analyzing factors indicating the reason why the product sold based on the information; and a display means for displaying the results of the analysis. 2. The information processing device according to 1, characterized in that the analysis means analyzes the factors using a plurality of methods and generates the reason why the product sold using a large-scale language model. 3. The information processing device according to any one of 1 to 2, characterized in that the analysis means comprises an identification means for identifying a product or service that deserves attention in the market based on the analysis results. 4. The information processing device according to any one of 1 to 3, characterized in that the analysis means takes information on the company's own product group and the product group of other companies, and market information as input, determines the success or failure of a product, and analyzes the factors of successful products of other companies and the differences with the company's own products. 5. The information processing device according to any one of 1 to 4, further comprising a hypothesis generation means for generating hypotheses for new product development based on the analysis results of the analysis means. 6. The information processing device according to any one of 1 to 5, characterized in that the acquisition means acquires image data and identifies a product from the image data. 7. 10. The information processing device according to any one of claims 1 to 6, wherein the acquisition means acquires a plurality of images including the product, and the analysis means performs an analysis based on the plurality of images and the time the plurality of images were taken. 8. The information processing device according to any one of claims 1 to 7, further comprising a design generation means that generates a product design based on the analysis results of the analysis means, wherein the design generation means comprises a text generation means that generates text describing the analysis results or the characteristics of the identified product using a machine learning model, and a label generation means that uses an image generation model with the generated text as input to generate a new product image. 9. The information processing device according to any one of claims 1 to 8, wherein the analysis means analyzes the factors using at least one of feature vector automatic design, large-scale language models, and genetic algorithms. 11. The information processing device according to any one of claims 1 to 9, wherein the analysis means analyzes the factors based on both quantitative and qualitative data.11. The information processing device according to 10, wherein the analysis means generates features from quantitative data and qualitative data by automatic feature design, analyzes at least one of the generated features and qualitative data with a large-scale language model, and optimizes the analysis results of the generated features and large-scale language model with a genetic algorithm. 12. The information processing device according to 3, wherein the identification means identifies a product or service of note in the market using at least one of sales analysis, sentiment analysis, trend analysis, and competitive analysis. 13. The information processing device according to 3, wherein the identification means identifies a product or service of note in the market using at least two of sales analysis, sentiment analysis, trend analysis, and competitive analysis. 14. The information processing device according to 3, 12, or 13, wherein the identification means analyzes at least one of the success factors and failure factors for the identified product or service. 15. The information processing device according to 14, wherein the identification means extracts commonalities of successful products or services. 16. The information processing device according to 15, wherein the identification means analyzes the correlation between the extracted commonalities and sales or customer satisfaction. 17. 18. The identification means is an information processing device according to any one of 14 to 16 that extracts commonalities between unsuccessful products or services. 19. The hypothesis generation means is an information processing device according to 5 that generates hypotheses including ideas for new products that are expected to be successful. 20. An information processing method in which a computer acquires information about a product, analyzes factors indicating why the product sold based on the information, and displays the results of the analysis. 20. A recording medium that records a program causing a computer to perform an acquisition step of acquiring information about a product, an analysis step of analyzing factors indicating why the product sold based on the information, and a display step of displaying the results of the analysis.

[0186] Some or all of the appendices 2 to 18, which are dependent on the information processing device described in appendice 1 above, may also be dependent on the information processing method in appendice 19 and the recording medium in appendice 20 in the same dependent relationship as between appendice 1 and appendices 2 to 18. Furthermore, some or all of the configurations described as appendices can be realized in various hardware, software, various recording means for recording software, or systems, without departing from the embodiments described above.

[0187] 10 Information Processing Device 110 Acquisition Unit 120 Analysis Unit 130 Display Unit 20 Information Processing Device 210 Acquisition Unit 220 Analysis Unit 221 Identification Unit 230 Display Unit 30 Information Processing Device 310 Acquisition Unit 320 Feature Generation Unit 330 Analysis Unit 331 Identification Unit 340 Display Unit 40 Information Processing Device 440 Hypothesis Generation Unit 101 Processor 102 Memory 103 Communication Interface 104 Program

Claims

1. An information processing device comprising: an acquisition means for acquiring information about a product; an analysis means for analyzing factors indicating the reason why the product was sold based on the information; and a display means for displaying the results of the analysis.

2. The information processing apparatus according to claim 1, characterized in that the analysis means analyzes the factors using multiple methods and generates reasons for the sale of the product using a large-scale language model.

3. The information processing apparatus according to claim 1 or 2, characterized in that the analysis means comprises a means for identifying a product or service of note in the market based on the results of the analysis.

4. The information processing device according to any one of claims 1 to 3, characterized in that the analysis means takes information on the company's own product group and the competitor's product group, as well as market information, as input to determine the success or failure of a product, and analyzes the factors contributing to the success of competitors' products and the differences between those products and the company's own products.

5. The information processing apparatus according to any one of claims 1 to 4, further comprising a hypothesis generation means for generating hypotheses for new product development based on the analysis results of the analysis means.

6. The information processing apparatus according to any one of claims 1 to 5, characterized in that the acquisition means acquires image data and identifies a product from the image data.

7. The information processing apparatus according to any one of claims 1 to 6, characterized in that the acquisition means acquires a plurality of images including the product, and the analysis means performs an analysis based on the plurality of images and the time when the plurality of images were taken.

8. The information processing apparatus according to any one of claims 1 to 7, further comprising a design generation means for generating a product design based on the analysis results of the analysis means, wherein the design generation means comprises a text generation means for generating text describing the analysis results or the characteristics of the identified product using a machine learning model, and a label generation means for generating a new product image using an image generation model with the generated text as input.

9. The information processing apparatus according to any one of claims 1 to 8, wherein the analysis means analyzes the factors using at least one of feature automatic design, a large-scale language model, and a genetic algorithm.

10. The information processing apparatus according to any one of claims 1 to 9, wherein the analysis means analyzes the factors based on both quantitative and qualitative data.

11. The information processing apparatus according to claim 10, wherein the analysis means generates features from quantitative data and qualitative data by automatic feature design, analyzes at least one of the generated features and qualitative data with a large-scale language model, and optimizes the analysis results of the generated features and the large-scale language model with a genetic algorithm.

12. The information processing apparatus according to claim 3, wherein the identifying means identifies a product or service of note in the market using at least one of sales analysis, sentiment analysis, trend analysis, and competitive analysis.

13. The information processing apparatus according to claim 3, wherein the identifying means identifies a product or service of note in the market using at least two of the following: sales analysis, sentiment analysis, trend analysis, and competitive analysis.

14. The information processing apparatus according to claim 3, 12, or 13, wherein the identifying means analyzes at least one of the success factors and failure factors for the identified goods or services.

15. The information processing apparatus according to claim 14, wherein the identifying means extracts commonalities of successful products or services.

16. The information processing apparatus according to claim 15, wherein the identifying means analyzes the correlation between the extracted commonalities and sales or customer satisfaction.

17. The information processing apparatus according to any one of claims 14 to 16, wherein the identifying means extracts commonalities between failed products or services.

18. The information processing apparatus according to claim 5, wherein the hypothesis generating means generates the hypothesis, which includes an idea for a new product that is considered likely to be successful.

19. An information processing method comprising: a computer acquiring information about a product; analyzing factors indicating the reason why the product was sold based on the information; and displaying the results of the analysis.

20. A recording medium that stores a program that causes a computer to perform an acquisition step of acquiring information about a product, an analysis step of analyzing factors indicating the reason why the product was sold based on the information, and a display step of displaying the results of the analysis.

Images