System and method for supporting analysis of sales activities

Abstract

To provide a sales activity analysis support system and method for analyzing sales activities at a further high level.SOLUTION: A sales activity analysis support system S includes: an accumulated information summarizing unit that receives a behavior history database for accumulating behavior information for every sales case, a case management database for storing case management information including stakeholder information, and identification information for identifying a sales case for which one or more input documents are created, acquires the case management information from the case management database, acquires behavior information relating to the sales case accumulated in the behavior history database when the input documents are received, and generates summary data; a context inference unit that generates context information representing a context until the input documents are received on the basis of the summary data and the input document; a behavior analysis unit that generates behavior information on the basis of the context information and the input documents, and records the generated behavior information in the behavior history database; and a sales activity analysis unit that performs analysis processing of sales activities on the basis of the behavior information accumulated in the behavior history database.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a technology for supporting the analysis of sales activities using document data. [Background technology]

[0002] Sales activities play an important role in corporate management. However, sales activities tend to be based on the personal know-how of each salesperson, making it difficult to manage sales activities efficiently and effectively and to formulate strategies. In addition, sales activities generally involve multiple and numerous projects being carried out in parallel, making it difficult for managers to grasp the overall picture of sales activities in real time.

[0003] Against this backdrop, development has progressed in technologies for managing and analyzing sales activities, as well as systems for centrally managing sales activities, such as CRM (Customer Relationship Management) and SFA (Sales Force Automation) (hereinafter also referred to as "CRM / SFA systems") (see, for example, Patent Documents 1 and 2). For example, CRM / SFA systems aggregate sales opportunity information (e.g., customer name, industry, order amount, lead time, etc.) and perform statistical analysis and data visualization along various axes. Furthermore, CRM / SFA systems can perform actions such as sending emails and setting up web conferences. By recording these actions, these systems enable the analysis and visualization of sales activities. However, existing technologies and systems are limited in the actions they can record, making it impossible to analyze and understand the details of activities, making it difficult to achieve advanced management, efficiency, and strategy formulation of sales activities. To enable more detailed analysis of activities, there is a need for more diverse and detailed recording of the actions of sales representatives and stakeholders.

[0004] One solution to this problem is to utilize document data (for example, Non-Patent Document 1). In CRM / SFA systems, document data that records various details about sales activities is registered during the sales process. By reading and understanding the document data, it is possible to understand what actions were taken by whom. In order to achieve advanced and efficient sales activity analysis based on document data that is accumulated daily, technology is needed that can comprehensively identify and record the actions of sales representatives and stakeholders from the content of the document data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-123183 [Patent Document 2] Japanese Patent Application Laid-Open No. 2009-157538 [Non-patent literature]

[0006] [Non-Patent Document 1] Yuta Hayashibe, Mamoru Komachi, Yuji Matsumoto, "Japanese Predicate-Argument Structure Analysis by Comparing Candidates Based on the Positional Relationship Between Predicates and Arguments," Natural Language Processing 21.1 (2014) pp.3-25 Summary of the Invention [Problem to be solved by the invention]

[0007] The technology described in Patent Document 1 relates to a sales support system that recommends actions with a high expected value for future profits based on the sales daily report history for a specified sales target. The sales support system uses reinforcement learning to learn how to recommend actions based on the "performed actions" and "action results" that accompany the sales daily reports. However, it is assumed that the "performed actions" are explicitly associated with the sales daily reports, and does not disclose a technology for identifying the content of actions from arbitrary document data. The content of the "performed actions" themselves is merely a keyword of a few characters, and the subject of the actions is limited to sales representatives.

[0008] The technology described in Patent Document 2 relates to a technique for plotting the progress of negotiations and other information as graphs on text data using clues such as phase names, customer names, contact person names, and product names in order to grasp the overall picture of business processes such as sales activities within a limited display area. However, this technique is limited to displaying analytical information on sales activities based on keywords present in the document data, and does not disclose a technique for identifying diverse and detailed behavioral content from the document data. Furthermore, it is not possible to aggregate and analyze the information identified from the document data across sales cases.

[0009] The technology described in Non-Patent Document 1 discloses a technology related to predicate-argument structure analysis that acquires from document data the relationships of who, what, and how from predicates in a sentence and their complements. However, the analysis results are based on the predicates written in the sentence, and it is not possible to analyze the actions of salespeople, etc., from daily reports, etc., which are not explicitly stated as predicates but can be determined self-evidently from the sentence. Furthermore, since it is not self-evident which position (sales representative, customer representative, etc.) the subject and object of the predicate take in sales activities, the disclosed technology misses out on important information when analyzing sales activities.

[0010] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a technology that can realize more advanced analysis of sales activities by comprehensively and automatically acquiring diverse and detailed behavioral content that could not be acquired using conventional technology based on the content of document data. [Means for solving the problem]

[0011] A sales activity analysis support system according to the present invention is a system for supporting the analysis of sales activities, and includes: a behavior history database that accumulates behavioral information indicating behavior in sales activities for each sales case; and a case management database that stores, for each sales case, case management information including information on stakeholders in each sales case. The system also includes an accumulated information summarization unit that, upon receiving one or more input documents related to the sales case and identification information that uniquely identifies the sales case for which the input documents were created, acquires the case management information related to the sales case from the case management database, acquires behavioral information related to the sales case that is accumulated in the behavior history database at the time the input documents are received, and generates summary data based on the acquired behavioral information and case management information; a process estimation unit that generates process information indicating the process up to the reception of the input documents based on the summary data and the input documents; a behavior analysis unit that newly generates zero or more pieces of behavioral information based on the process information and the input documents and records the generated behavioral information in the behavior history database; and a sales behavior analysis unit that performs analytical processing of the sales activities based on the behavioral information accumulated in the behavior history database.

[0012] Other problems and solutions disclosed in the present application will be made clear in the detailed description and drawings. [Effects of the Invention]

[0013] According to the present invention, more advanced analysis of sales activities can be realized by comprehensively and automatically acquiring diverse and detailed behavioral details based on the contents of document data, which could not be acquired by conventional techniques. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a diagram illustrating an example of a hardware configuration of a sales activity analysis support system according to first to third embodiments. [Figure 2] FIG. 2 is a diagram illustrating an example of functional blocks of the sales activity analysis support system according to the first embodiment. [Figure 3]FIG. 2 is a diagram illustrating an example of a flow of processing executed by the sales activity analysis support system according to the first embodiment. [Figure 4] 10 is a flowchart showing an example of the flow of stored information summarization processing in the first and third embodiments. [Figure 5] FIG. 2 is a diagram showing an example of the configuration of input data in the first to third embodiments. [Figure 6] FIG. 2 is a diagram showing an example of the configuration of case management information in the first to third embodiments. [Figure 7] FIG. 2 is a diagram showing an example of the configuration of a behavior information list in the first to third embodiments. [Figure 8] FIG. 2 is a diagram showing an example of the structure of summary data in Examples 1 to 3. [Figure 9] 10 is a flowchart showing an example of the flow of a process of inferring circumstances in the first and third embodiments. [Figure 10] FIG. 10 is a diagram illustrating an example of the configuration of context data in the first and third embodiments. [Figure 11] 1 is a flowchart showing an example of the flow of a behavior analysis process in the first to third embodiments. [Figure 12] FIG. 10 is a diagram showing an example of the configuration of a behavior analysis result in Examples 1 to 3. [Figure 13] FIG. 10 is a diagram showing an example of the configuration of a new behavior information list in the first to third embodiments. [Figure 14] 10 is a flowchart showing an example of the flow of a sales behavior analysis process in the first and second embodiments. [Figure 15] FIG. 2 is a diagram showing an example of the configuration of a behavior information list in the first to third embodiments. [Figure 16] FIG. 2 is a diagram showing an example of the configuration of a behavior information list for display in Examples 1 and 2. [Figure 17] FIG. 10 is a diagram showing an example of the configuration of a calendar screen in the first and second embodiments. [Figure 18] FIG. 2 is a diagram showing an example of functional blocks of a sales activity analysis support system according to second and third embodiments. [Figure 19] FIG. 11 is a diagram illustrating an example of a flow of processing executed by the sales activity analysis support system in the second embodiment. [Figure 20] 10 is a flowchart illustrating an example of the flow of a stored information summarization process according to the second embodiment. [Figure 21] FIG. 10 is a diagram showing an example of the configuration of context data in the second and third embodiments. [Figure 22] 10 is a flowchart showing an example of the flow of a process of inferring circumstances in the second embodiment. [Figure 23] FIG. 11 is a diagram illustrating an example of the configuration of a case management information list according to the second embodiment. [Figure 24] FIG. 11 is a diagram illustrating an example of the configuration of a cross-case behavior information list according to the second embodiment. [Figure 25] FIG. 10 is a diagram illustrating an example of a configuration of similar behavior information data according to the second embodiment. [Figure 26] FIG. 11 is a diagram illustrating an example of the configuration of a history information candidate list in the second embodiment. [Figure 27] 13 is a flowchart showing an example of the flow of a sales behavior analysis process in the third embodiment. [Figure 28] FIG. 11 is a diagram showing an example of the configuration of an analysis candidate behavior information list in the third embodiment. [Figure 29] FIG. 11 is a diagram showing an example of the configuration of a behavioral information list for analysis in the third embodiment. [Figure 30] FIG. 11 is a diagram showing an example of the configuration of an analysis screen in the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, the embodiments will be described in detail with reference to the drawings. However, the present invention is not limited to the description of the following examples. Examples in which the specific configuration is modified are also included within the scope of the idea or purpose of the present invention. For example, the following examples are intended to explain the present invention in detail, and are not necessarily limited to those having all the configurations included in the description.

[0016] In the configuration of the invention described below, the same parts and / or elements, or parts and / or elements having similar functions, will be denoted by the same symbols in different drawings, and duplicated explanations may be omitted.

[0017] Furthermore, when there are multiple identical parts and / or elements, or parts and / or elements with similar functions, the same reference numerals may be used with different subscripts to distinguish between the multiple parts and / or elements. On the other hand, when there is no need to distinguish between the multiple parts and / or elements, the subscripts may be omitted.

[0018] The designations "first," "second," "third," etc. in this specification are used to identify components and do not necessarily limit the number, order, or content thereof. Furthermore, numbers used to identify components are used in different contexts, and numbers used in one context do not necessarily indicate the same configuration in another context. Furthermore, this does not prevent a component identified by a certain number from also serving the function of a component identified by another number.

[0019] To facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in this specification and / or the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in this specification and / or the drawings.

[0020] As used herein, elements referred to in the singular are intended to include the plural unless the context clearly indicates otherwise.

[0021] Also, in the following description, an "interface apparatus" may refer to one or more interface devices, which may be at least one of the following: One or more I / O (Input / Output) interface devices. The I / O (Input / Output) interface devices are interface devices to at least one of the I / O device and a remote display computer. The I / O interface device to the display computer may be a communications interface device. The at least one I / O device may be a user interface device, for example, either an input interface device such as a keyboard and pointing device, or an output interface device such as a display device. One or more communication interface devices. The one or more communication interface devices may be one or more communication interface devices of the same type (for example, one or more NICs (Network Interface Cards)) or two or more communication interface devices of different types (for example, a NIC and an HBA (Host Bus Adapter)). Note that the network accessed by the communication interface device for communication may be the Internet, a LAN (Local Area Network), a WAN (Wide Area Network), a mobile phone network, or the like, but is not limited to these.

[0022] In the following description, "memory" refers to one or more memory devices, which are an example of one or more storage devices, and may typically be a primary storage device. At least one memory device in the memory may be a volatile memory device or a non-volatile memory device.

[0023] In the following description, an "auxiliary storage device" may be one or more persistent storage devices, which are an example of one or more storage devices. The auxiliary storage device may typically be a non-volatile storage device, specifically, for example, a hard disk drive (HDD), a solid state drive (SSD), or a non-volatile memory express (NVMe) drive.

[0024] In the following description, the term "storage device" may refer to at least one memory, including a memory and an auxiliary storage device.

[0025] Furthermore, in the following description, a "processor" which is an arithmetic unit may be one or more processor devices. The at least one processor device may typically be a microprocessor device such as a CPU (Central Processing Unit), but may also include other types of processor devices such as a GPU (Graphics Processing Unit). The at least one processor device may be a single-core or multi-core. The at least one processor device may be a processor core. The at least one processor device may also be a processor device in a broader sense, such as a hardware circuit that performs part or all of the processing (for example, an FPGA (Field-Programmable Gate Array), a CPLD (Complex Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit)).

[0026] In the following description, information that provides an output in response to an input may be described using expressions such as "xxx database" or "xxx table." However, this information may be data of any structure (for example, structured data or unstructured data), or may be a learning model such as a neural network, genetic algorithm, or random forest that generates an output in response to an input. Therefore, "xxx database" or "xxx table" can be rephrased as "xxx information." In the following description, the structure of each database or table is an example, and one database or table may be divided into two or more databases or tables, or two or more databases or tables may all or partly be one database or table.

[0027] In the following description, processing may be described using a "program" as the subject; however, since a program is executed by a processor to perform a predetermined process using a storage device and / or an interface device, etc., as appropriate, the subject of the process may also be the processor (or a device such as a controller having the processor). A program may be installed in a device such as a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable (e.g., non-transitory) recording medium. In the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0028] Furthermore, in the following description, the "sales activity analysis support system" may be a system (e.g., a cloud computing system) realized on a group of physical computing resources (e.g., a cloud infrastructure), or may be a system (e.g., an on-premise system) configured with one or more physical computers. When the sales activity analysis support system "displays" the display information, it may mean that the display information is displayed on a display device possessed by the computer, or that the computer transmits the display information to a display computer (in the latter case, the display information is displayed by the display computer). [Example]

[0029] <System configuration example> First, an example of the configuration of the sales activity analysis support system S will be described with reference to FIGS.

[0030] (Example of the configuration of the sales activity analysis support system S) The sales activity analysis support system S in Example 1 is a computer system that enables detailed and efficient review of the actions of sales representatives engaged in sales as part of the business of a company, etc., thereby supporting the analysis of sales activities by the sales representatives, and is realized by a computer device or server device having the respective configurations described below.

[0031] Fig. 1 is a diagram showing an example of the hardware configuration of an information processing device J, which is a computer device or a server device, constituting the sales activity analysis support system S in the first embodiment (and in the second and third embodiments). Fig. 2 and Fig. 3 are diagrams showing examples of functional blocks of the sales activity analysis support system S in the first embodiment.

[0032] The information processing device J constituting the sales activity analysis support system S is configured to be capable of data communication with other computer devices, server devices, etc. (hereinafter also referred to as "other devices") via a network 50. For example, the information processing device J acquires an input document 101 from another device via the network 50 (details will be described later). Note that the information processing device J and the network 50 are connected by wire via well-known communication equipment (not shown), but may also be connected wirelessly. In this case, the other device and the network 50 may be connected by wire or wirelessly via well-known communication equipment (not shown).

[0033] Furthermore, user terminals (not shown), such as laptop PCs, tablets, and smartphones owned by operators who are users of the sales activity analysis support system S, may be connected to the information processing device J constituting the sales activity analysis support system S so as to enable mutual data communication via an appropriate network 50, such as the Internet or a dedicated line. In this case, each user terminal and the network 50 may be connected wirelessly or by wire.

[0034] 1 and 2, the sales activity analysis support system S has been described as being composed of one information processing device J. However, for example, the sales activity analysis support system S may be composed of a plurality of information processing devices.

[0035] In addition, in this embodiment, the information processing device J constituting the sales activity analysis support system S and various other devices such as user terminals and other devices have been described as being separate devices. However, the information processing device J constituting the sales activity analysis support system S and various other devices may be configured as the same device. In this case, the sales activity analysis support system S may be configured as, for example, a computer system including some or all of these various devices. Furthermore, for example, the sales activity analysis support system S may be configured to include some or all of the functions performed by these various devices.

[0036] (Example of hardware configuration for the Sales Activity Analysis Support System S) Next, an example of the hardware configuration of the information processing device J that constitutes the sales activity analysis support system S will be described with reference to FIG.

[0037] 1, this sales activity analysis support system S is realized by a computer having at least a storage device including a memory 2 and an auxiliary storage device 3, an interface device including at least a communication interface 4, and a processor 1 connected thereto. In this sales activity analysis support system S, the interface device may also include an input interface 5 and / or an output interface 6.

[0038] The following explanation will be given assuming that the sales activity analysis support system S is realized by a single general-purpose computer device having one or more processors 1, one or more memories 2, one or more auxiliary storage devices 3, one or more communication interfaces (hereinafter also referred to as "communication I / F") 4, one or more input interfaces (hereinafter also referred to as "input I / F") 5, one or more output interfaces (hereinafter also referred to as "output I / F") 6, and wired or wireless communication lines connecting them.

[0039] The auxiliary storage device 3 is an auxiliary storage device made up of a nonvolatile storage element such as a flash memory. Specific examples of the auxiliary storage device 3 include a solid state drive (SSD) and a hard disk drive (HDD). The auxiliary storage device 3 stores at least a sales activity analysis support program (not shown). The sales activity analysis support program is a computer program for implementing the functions required for the sales activity analysis support system S.

[0040] That is, when the sales activity analysis support program is executed by the processor 1, the functions of the respective functional units of the sales activity analysis support system S, such as the accumulated information summarizing unit 102, the process inferring unit 103, the behavior analysis unit 104, and the sales behavior analysis unit 105, which will be described later, are realized. In other words, when the sales activity analysis support program is executed by the processor 1, various processes are performed, including a process relating to the summarization of accumulated information (hereinafter also referred to as "accumulated information summarizing process"), a process relating to the inference of process (hereinafter also referred to as "process inferring process"), a process relating to the analysis of behavior (hereinafter also referred to as "behavior analysis process"), and a process relating to the analysis of sales behavior (hereinafter also referred to as "sales behavior analysis process"), which will be described later with reference to Figures 4 to 17.

[0041] The sales activity analysis support program is provided to the sales activity analysis support system S from various removable media such as a CD-ROM or flash memory or via the network 50, and is stored in the non-volatile auxiliary storage device 3, which is a non-temporary storage medium. Therefore, it is preferable that the sales activity analysis support system S has an interface for reading data from the removable media.

[0042] The sales activity analysis support program may also be installed from a program source. The program source may be, for example, a computer on which the program is distributed or a computer-readable recording medium. The sales activity analysis support program may also be configured by a device driver, an operating system, various application programs located at higher levels than these, and a library that provides common functions to these programs. Furthermore, two or more programs may be realized as one sales activity analysis support program, or one sales activity analysis support program may be realized as two or more programs.

[0043] The memory 2 is a primary storage device mainly consisting of volatile storage elements such as RAM (Random Access Memory). The memory 2 also includes a ROM consisting of nonvolatile storage elements. The ROM stores unchanging programs (e.g., BIOS). The memory 2 temporarily stores data representing various information read from the auxiliary storage device 3 and various data acquired via the communication interface 4 and / or input interface 5.

[0044] The processor 1 is a processor device such as a CPU (Central Processing Unit) and various co-processors. The processor 1 loads various computer programs, including a sales activity analysis support program, into the memory 2 and executes them, thereby performing overall control of the sales activity analysis support system S itself, and also controls the control unit 11 that performs various processes such as calculation processing and judgment processing.

[0045] The interface device includes a communication interface 4 that controls a communication unit 14 described below, an input interface 5 that controls an input unit 51 described below, and an output interface 6 that controls an output unit 61 described below.

[0046] The communication interface 4 is a network interface device that controls communication with other devices in accordance with a predetermined protocol.

[0047] The input interface 5 is an interface to which input devices such as a keyboard 7, a mouse 8, a touch panel, etc. are connected and which receives input from an operator.

[0048] The output interface 6 is an interface to which various display devices 9 such as a liquid crystal display or a touch screen, or an output device such as a printer (not shown), are connected, and which outputs the results of program execution in a format that can be viewed by an operator.

[0049] The sales activity analysis support system S may be an independent device or may be an embedded device.

[0050] (Example of functional blocks of the sales activity analysis support system S) Next, an example of various functional blocks included in the sales activity analysis support system S in the first embodiment will be described with reference to Figures 2 and 3. Note that each block described below does not represent a hardware configuration, but represents a functional block.

[0051] The sales activity analysis support system S is mainly composed of functional blocks: a control unit 11 realized by the aforementioned processor 1; a memory unit 13 realized by the aforementioned memory devices (2, 3); a communication unit 14 realized by the aforementioned communication interface 4; and a user interface unit 15 realized by the aforementioned input interface 5 and output interface 6.

[0052] The control unit 11 executes various data processing operations based on the programs and data stored in the storage unit 13 and the data acquired by the communication unit 14. The control unit 11 also functions as an interface between the storage unit 13 and the communication unit 14.

[0053] The control unit 11 has the following functional blocks: an accumulated information summarizing unit 102, a process inferring unit 103, a behavior analyzing unit 104, and a sales behavior analyzing unit 105, as illustrated in FIGS.

[0054] The stored information summarizing unit 102 executes a stored information summarizing process, the details of which will be described later with reference to FIGS.

[0055] The process inference unit 103 executes a process inference process, the details of which will be described later with reference to FIGS.

[0056] The behavior analysis unit 104 executes a behavior analysis process, the details of which will be described later with reference to Figs.

[0057] The sales behavior analysis unit 105 executes a sales behavior analysis process, the details of which will be described later with reference to FIGS.

[0058] The control unit 11 is configured using the processor 1, and can realize these functional blocks by executing the sales activity analysis support program described above. Note that the control unit 11 may be configured using a logic circuit such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit) instead of the processor 1. The control unit 11 may also be configured by combining the processor 1 and a logic circuit.

[0059] The memory unit 13 is configured using a storage device, for example, consisting of a memory 2 and an auxiliary storage device 3, and stores a program that supplies various processing commands to the control unit 11, and data representing various information used in the processing executed by the control unit 11.

[0060] The storage unit 13 stores at least an action history database 106 and a case management database 107, as illustrated in FIGS.

[0061] The behavioral history database 106 is a database that stores information (hereinafter simply referred to as "behavioral information") representing the behavioral history of salespeople and the like, acquired based on the input document 101, which is document data that records various details related to sales activities. The sales activity analysis support system S utilizes the behavioral information accumulated in this behavioral history database 106 to realize detailed and advanced sales activity analysis.

[0062] The case management database 107 is a database that stores various information for managing cases (hereinafter also simply referred to as "case management information").

[0063] By reading and writing data representing various pieces of information managed by these databases to the memory unit 13, the control unit 11 can execute various processes including the aforementioned accumulated information summarization process (described in detail below in relation to Figures 4 to 8), background inference process (described in detail below in relation to Figures 8 to 10), behavior analysis process (described in detail below in relation to Figures 7 and Figures 10 to 13), and sales behavior analysis process (described in detail below in relation to Figures 14 to 17).

[0064] The communication unit 14 is responsible for communication processing with user terminals and other devices via the network 50. The communication unit 14 is configured using, for example, a NIC (Network Interface Card) or an HBA (Host Bus Adapter).

[0065] The user interface unit 15 includes the functional blocks of an input unit 51 and an output unit 61.

[0066] The input unit 51 is responsible for input-related processing, such as accepting input operations from the user, among other processes related to the user interface. The input unit 51 is configured using input devices such as the keyboard 7, mouse 8, touch panel, etc., and detects various operations by the user.

[0067] Among the processes related to the user interface, the output unit 61 is responsible for output-related processes such as displaying various screens and outputting audio on the display device 9. The output unit 61 is configured using various display devices 9 such as a liquid crystal display or a touch screen.

[0068] It should be noted that the inclusion of the input unit 51 and / or the output unit 61 is not essential, for example, when remotely logging in to the sales activity analysis support system S from another external device such as a user terminal, or when receiving input information from an external device or providing output information to an external device via the communication interface 4. In this case, the sales activity analysis support system S may have a web server function and receive access from an external device using a predetermined protocol.

[0069] In other words, each component of the sales activity analysis support system S is realized by hardware including a processor 1, storage devices such as a memory 2 and an auxiliary storage device 3, wired or wireless communication lines and interface devices (4, 5, 6) that connect them, and software stored in the storage devices (2, 3) that supplies processing instructions to the arithmetic unit (processor 1).

[0070] The above description of the functions of the sales activity analysis support system S has been given assuming that each function of the sales activity analysis support system S is implemented integrally by a single computer. However, each function may be implemented by multiple interconnected computers and / or server devices. Furthermore, the sales activity analysis support system S may be configured to include a general-purpose computer such as a laptop PC with a web browser installed thereon, or may be configured to include a web server and various portable devices.

[0071] The sales activity analysis support system S is a computer system configured on one physical computer or on multiple logically or physically configured computers, and may operate on a virtual computer constructed on multiple physical computer resources. For example, functional units such as the accumulated information summarization unit 102, the process inference unit 103, the behavior analysis unit 104, and the sales behavior analysis unit 105 may each operate on a separate physical or logical computer, or multiple units may be combined to operate on a single physical or logical computer.

[0072] Furthermore, the above description of each function is merely an example, and multiple functions may be combined into one function, or one function may be divided into multiple functions.

[0073] Furthermore, the sales activity analysis support system S may have other functions in addition to the above functions. For example, as described above, the sales activity analysis support system S may be configured to include some of the various functions of other devices.

[0074] <Processing flow example> Next, the above-mentioned processes executed by the sales activity analysis support system S in the first embodiment will be described with reference to FIGS.

[0075] (Stored information summary processing) FIG. 4 is a flowchart showing an example of the flow of the stored information summarizing process in the first embodiment (and the third embodiment).

[0076] When the control unit 11 of the sales activity analysis support system S receives the input document 302 and the case identification information 305 that uniquely identifies the sales case for which the input document 302 was created, the control unit 11 executes the stored information summarization process described above, in which the stored information summarization unit 102 retrieves case management information 401 related to the sales case from the case management database 107 and retrieves behavior information 502 related to the sales case that has already been stored in the behavior history database 106, and generates summary data 701 based on the extracted behavior information 502 and the case management information 401. Details of this stored information summarization process will be described using Figures 4 to 8.

[0077] In step S401, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 accepts input data 301 consisting of a pair of an input document 302 and case identification information 305. The case identification information 305 may be in any data format as long as it can uniquely identify the case. The input document 302 consists of a writing date 303 and text 304. However, the writing date 303 does not have to be the date the input document 302 was written, as long as it is a date linked to the input document 302. For example, it may be the date the input document 302 was received or the date the activity described in the text 304 was performed. Furthermore, the text 304 can be any string of characters consisting of one or more characters. For example, various text data such as reports, proposals, minutes, memos, emails, conversation transcripts, chat histories, program logs, and in-house system log messages can be accepted as the text 304 of the input document 302. Multiple pieces of input data 301 can be accepted in step S401. When multiple pieces of input data 301 are received, the processing from step S402 onwards is performed for each piece of input data 301. In the above-mentioned situation, the process estimation unit 103 and the behavior analysis unit 104 also perform processing on each piece of summary data 701 output by the accumulated information summarization unit 102 based on the multiple pieces of input data 301. This results in the acquisition of input data 301 consisting of pairs of input documents 302 and case identification information 305. When the processing in step S401 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S402.

[0078] In step S402, the control unit 11 of the sales activity analysis support system S executes a process of acquiring, from the case management database 107, case management information 401 of a case that matches the case identification information 305 described above, using the accumulated information summarizing unit 102. Here, the case management information 401 is data that stores values ​​indicating characteristics of a sales case, item by item. In the first embodiment, the case management information 401 acquired in step S402 includes information on a case ID 402 that uniquely identifies the case, a sales representative 403 that indicates the company's sales representative for the case, a customer name 404 that indicates the name of the customer targeted by the case, and a customer representative 405 that indicates the sales representative for the customer for the case. However, the information on each case stored in the case management database 107 may have more items than those acquired in step S402. Furthermore, the case management information 401 acquired in step S402 may include information on other stakeholders related to the case, and more detailed profile information, such as the department to which the sales representative 403 and the customer representative 405 belong. In addition, the case management information 401 acquired in step S402 may include various other information describing the case, such as the case's project structure, type, customer industry, size, expected order amount, case creation date, expected order date, lead time, importance, etc. This allows the case management information 401 of the case that matches the above-mentioned case identification information 305 to be acquired from the case management database 107. Upon completing the processing in step S402, the control unit 11 of the sales activity analysis support system S proceeds to step S403.

[0079] In step S403, the control unit 11 of the sales activity analysis support system S executes processing for initially creating summary text 702 using the stored information summarizing unit 102. The summary text 702 is a string of any one or more characters. In the first embodiment, the summary text 702 created in step S403 is "You are reading the activity details registered in the database by the sales representative for the ongoing sales case." In this way, the summary text 702 is initially created. Upon completing the processing in step S403, the control unit 11 of the sales activity analysis support system S proceeds to step S404.

[0080] In step S404, the control unit 11 of the sales activity analysis support system S executes a process of adding the summary text 702 using the case management information 401 acquired in step S402, using the stored information summarizing unit 102. In the first embodiment, the person in charge 403, the customer name 404, and the customer person in charge 405 in the case management information 401 are applied to a template such as "The following is the information about the sales case. Person in charge: {value of person in charge 403} Customer: {value of customer name 404} Customer person in charge: {value of customer person in charge 405}" and added to the summary text 702. However, a different template may also be used. In this way, the summary text 702 is added using the case management information 401. When the process in step S404 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S405.

[0081] In step S405, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 acquires, from the action history database 106, action information 502 having the same case ID 503 as the case identification information 305 of the input data 301 as an action information list 501. The acquired action information 502 is action information 502 that has already been accumulated in the action history database 106. The action information 502 is information describing the actions of sales representatives, stakeholders, etc., related to the sales case, and is accumulated in the action history database 106 by being generated by the action analysis unit 104 or manually registered. In the first embodiment, the behavior information 502 acquired in step S405 includes a case ID 503 indicating the case identification information 305 of the corresponding case, an action time 504 indicating the time when the behavior was performed, a subject 505 indicating the subject of the behavior, a subject position 506 indicating the subject's position in sales activities, an action content 507 explaining the content of the behavior, and a recording date 508 indicating the date on which the behavior information 502 was recorded. However, each piece of information constituting the behavior information 502 may take any form. For example, the action time 504 may be a character string specifying an arbitrary interval on a time axis rather than a point in time that clearly identifies the time. For example, the subject 505 does not necessarily have to clearly indicate a person's name. For example, the action content 507 may store any information as long as it is a character string of one or more characters. In addition, the behavior information 502 acquired in step S405 may include other additional information related to the behavior. For example, it may include the name of the department to which the subject 505 belongs, a character string indicating the object of the behavior, a character string indicating the intention of the behavior, etc. Note that the behavior history database 106 stores behavior information 502 up to the point when the behavior was last confirmed, but the behavior information list 501 acquired in step S405 does not need to include all of the behavior information 502 related to the case identification information 305 stored in the behavior history database 106. For example, a process of selecting only a specific behavior content 507 or a process of selecting only a specific behavior time 504 may be inserted. In the first embodiment, of the behavior information 502 having the same case ID 503 as the case identification information 305 of the input data 301 stored in the behavior history database 106, only the behavior information 502 whose recording date 508 is a date earlier than the entry date 303 of the input data 301 is acquired as the behavior information list 501.As a result, the behavior information 502 having the same case ID 503 as the case identification information 305 of the input data 301 is obtained from the behavior history database 106 as the behavior information list 501. When the processing in step S405 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S406.

[0082] In step S406, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 determines whether the behavior information list 501 acquired in step S405 is composed of one or more pieces of behavior information 502. If it is determined in step S406 that the behavior information list 501 is composed of one or more pieces of behavior information 502 (step S406: YES), the process proceeds to step S407. On the other hand, if it is determined in step S406 that the behavior information list 501 is not composed of one or more pieces of behavior information 502 (step S406: NO), the process proceeds to step S410.

[0083] In step S407, the control unit 11 of the sales activity analysis support system S causes the stored information summarizing unit 102 to start loop processing for each piece of behavior information 502 that makes up the behavior information list 501. This starts loop processing for each piece of behavior information 502 that makes up the behavior information list 501. When the processing in step S407 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S408.

[0084] In step S408, the control unit 11 of the sales activity analysis support system S executes processing to add to the summary text 702 using the accumulated information summarizing unit 102 based on the information stored in the behavioral information 502. In the first embodiment, the behavioral time 504 and the behavioral content 507 of the behavioral information 502 are applied to a template of "{value of the behavioral time 504} {summarized value of the subject's position 506 and the behavioral content 507}" to create added text, and the added text is added to the summary text 702. Summarizing the behavioral content 507 refers to, for example, converting a character string combining the subject's position 506 and the behavioral content 507 into a character string with fewer characters. The summarization processing may be performed based on pre-designed rules, or may be performed using a statistical model, machine learning, deep learning model, or a combination of these. However, summarization is not required. A template different from the above template may also be used. As a result, the summary text 702 is appended based on the information held in the behavior information 502. When the control unit 11 of the sales activity analysis support system S completes the process in step S408, the process proceeds to step S409.

[0085] In step S409, the control unit 11 of the sales activity analysis support system S causes the stored information summarizing unit 102 to perform the above processing on all of the behavior information 502 that constitutes the behavior information list 501, and then ends the loop processing.

[0086] In step S410, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 combines the summary text 702, the input document 302, and the case identification information 305 of the input document to generate summary data 701. In the first embodiment, the summary data 701 does not include the past behavior information 707. In some embodiments, the summary data 701 may have a different data structure. As a result, the summary text 702, the input document 302, and the case identification information 305 of the input document are combined to generate the summary data 701. When the process in step S410 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S411.

[0087] In step S411, the control unit 11 of the sales activity analysis support system S causes the accumulated information summarizing unit 102 to execute processing to output the summary data 701. This outputs the summary data 701. Upon completing the processing in step S411, the control unit 11 of the sales activity analysis support system S ends the accumulated information summarizing processing shown in the flowchart of FIG.

[0088] (History inference processing) FIG. 9 is a flowchart showing an example of the flow of the process of inferring circumstances in the first embodiment (and the third embodiment).

[0089] The control unit 11 of the sales activity analysis support system S executes the process of inferring the process described above by the process inference unit 103, which receives the summary data 701 output in step S411 of Fig. 4 and generates process information 1009 indicating the process up to the reception of the input data 301. Details of this process inferring process will be described with reference to Figs. 8 to 10.

[0090] In step S901, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 receives the summary data 701 output in step S411 of Fig. 4. This acquires the summary data 701. When the process in step S901 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S902.

[0091] In step S902, the control unit 11 of the sales activity analysis support system S executes a process of generating the inference instruction text 1008 using the input document 703 of the summary data 701 by the process inference unit 103. In the first embodiment, the inference instruction text 1008 is generated by applying the description date 704 and the text 705 of the input document 703 to a template that reads, "The following is a document registered in {the value of the description date 704} regarding this business case. What is this document? How was it written? Please infer the circumstances. Document registered in {the value of the description date 704}: {the value of the text 705}." However, a different template may also be used. In this way, the inference instruction text 1008 is generated using the input document 703 of the summary data 701. After completing the process in step S902, the control unit 11 of the sales activity analysis support system S proceeds to step S903.

[0092] In step S903, the control unit 11 of the sales activity analysis support system S causes the process estimation unit 103 to perform an inference process using the summary data 701 and the inference instruction text 1008, thereby generating process history information 1009. In the first embodiment, the summary text 702 of the summary data 701 and the inference instruction text 1008 are sequentially concatenated and provided as input to a content-generating deep learning model. The output of the content-generating deep learning model is obtained as process history information 1009. In the first embodiment, the information described in the process history information 1009 is the process leading up to the reception of new input data 301, and this process is inferred in consideration of the case background information based on the case management information 401 expressed in the summary text 702 and the brief history information of the behavior based on the behavior information 502 already stored in the behavior history database 106 expressed in the summary text 702. However, the model used to generate the history information 1009 need not be based on a deep learning model, but may be based on a statistical model or a machine learning model. As a result, an inference process is performed using the summary data 701 and the inference instruction text 1008 to generate the history information 1009. When the process in step S903 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S904.

[0093] In step S904, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 combines the summary data 701 acquired in step S901, the inference instruction text 1008 created in step S902, and the process information 1009 generated in step S903 to create context data 1001. As a result, the summary data 701, the inference instruction text 1008, and the process information 1009 are combined to create the context data 1001. When the process in step S904 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S905.

[0094] In step S905, the control unit 11 of the sales activity analysis support system S causes the process inference unit 103 to execute processing to output the context data 1001 created in step S904. This outputs the context data 1001. Upon completing the processing in step S905, the control unit 11 of the sales activity analysis support system S ends the process inference processing shown in the flowchart of FIG.

[0095] (Behavioral analysis processing) FIG. 11 is a flowchart showing an example of the flow of the behavior analysis process in the first embodiment (and the second and third embodiments).

[0096] The control unit 11 of the sales activity analysis support system S executes the above-mentioned behavior analysis process by the behavior analysis unit 104, which receives the context data 1001 output in step S906 of Fig. 9, generates zero or more pieces of behavior information as new behavior information 1402 based on the contents of the context data 1001, and records the generated new behavior information 1402 in the behavior history database 106. Details of this behavior analysis process will be described with reference to Fig. 7 and Figs. 10 to 13.

[0097] In step S1101, the control unit 11 of the sales activity analysis support system S executes processing to receive the context data 1001 output by the behavior analysis unit 104 in step S906 of Fig. 9. This acquires the context data 1001. When the processing in step S1101 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1102.

[0098] In step S1102, the control unit 11 of the sales activity analysis support system S executes processing for creating an analysis instruction text consisting of a character string of one or more characters using the behavior analysis unit 104. The analysis instruction text may be a uniform sentence, or may be a sentence created by applying the values ​​of stored data to an arbitrary template, or the analysis instruction text may be created using a statistical model, machine learning model, or deep learning model that outputs a character string from stored data. For example, the analysis instruction text may be created using a deep learning model that inputs text 1005 of context data 1001 and outputs analysis instruction text. In Example 1, a uniform sentence that is not dependent on the content of the context data 1001 is created: "Based on the results of the above study, please infer all of the actions of each person that can be considered from the content of the above document, and list who (subject), to whom (object), and what they did, for each action. In addition, please also tell us the subject's position in the sales activity (sales representative, customer, third party, etc.). The output should be a dictionary listing for each action, with the subject name, subject position, object name, action time, and action content as keys." This is used as the analysis instruction text. Upon completing the processing in step S1102, the control unit 11 of the sales activity analysis support system S proceeds to step S1103.

[0099] In step S1103, the control unit 11 of the sales activity analysis support system S executes a process in which the behavior analysis unit 104 creates text data, which is a character string, using the context data 1001 acquired in step S1101, and generates a new behavior information list 1401 consisting of zero or more new behavior information items 1402 based on the text data. In the first embodiment, the new behavior information 1402 includes a case ID 1403, a behavior time 1404, a subject 1405, the subject's position 1406, a behavior content 1407, and a recording date 1408. However, the new behavior information 1402 may include items other than those described above. For example, the new behavior information 1402 may include information such as an object and the object's position. Furthermore, depending on the embodiment, the new behavior information 1402 may not include items other than the case ID 1403, the recording date 1408, and the behavior content 1407. In the first embodiment, the summary text 1002, the inference instruction text 1008, the context information 1009, and the analysis instruction text are sequentially concatenated and provided as input to a content-generating deep learning model. The output of the content-generating deep learning model is obtained as analysis result text 1302. However, the model used to generate the analysis result text 1302 may not be based on a deep learning model, but may be based on a statistical model or a machine learning model. Once the analysis result text 1302 is obtained from the content-generating deep learning model, the analysis result text 1302 is analyzed using regular expression-based string matching and converted into zero or more pieces of new behavior information 1402. For example, by writing a regular expression that detects strings starting with "{" and ending with "}" and having one or more patterns of "{any string}":"{any string}"", sections in the analysis result text 1302 that describe the content of each piece of new behavior information 1402 are identified. In each section of the behavior information described above, a portion that matches the pattern "{any character string}":"{any character string}" is detected, and the character string enclosed in """ is obtained to obtain the behavior time 1404, subject 1405, subject's position 1406, and behavior content 1407 of the new behavior information 1402. The case identification information 1006 of the context data 1001 is added as the case ID 1403 to all the obtained new behavior information 1402.In addition, the description date 1004 of the context data 1001 is added as the recording date 1408 to all of the acquired new behavior information 1402. However, the date added as the recording date 1408 may be the date on which the accumulated information summarizing unit 102 received the input document 1003. A new behavior information list 1401 is created by arranging the created new behavior information 1402. Note that the above-mentioned process of creating the new behavior information list 1401 from the analysis result text 1302 may be performed based on another rule-based method, or may be performed using a statistical model, machine learning, deep learning model, or the like, or a combination thereof. For example, a deep learning model may be used that receives as input a character string obtained by sequentially concatenating the summary text 1002, text 1005, and history information 1009 of the context data 1001, and directly outputs a list of new behavior information 1402. Furthermore, if the analysis result text 1302 does not describe convertible behavior information, the new behavior information list 1401 may hold 0 new behavior information 1402. As a result, the new behavior information list 1401 is generated based on the text data created using the context data 1001. When the control unit 11 of the sales activity analysis support system S completes the process in step S1103, the control unit 11 proceeds to step S1104.

[0100] In step S1104, the control unit 11 of the sales activity analysis support system S executes a process in which the behavior analysis unit 104 determines whether the new behavior information list 1401 acquired in step S1103 is composed of one or more pieces of new behavior information 1402. If it is determined in step S1104 that the new behavior information list 1401 is composed of one or more pieces of new behavior information 1402 (step S1104: YES), the process proceeds to step S1105. On the other hand, if it is determined in step S1104 that the new behavior information list 1401 is not composed of one or more pieces of new behavior information 1402 (step S1104: NO), the behavior analysis process shown in the flowchart of FIG. 11 is terminated.

[0101] In step S1105, the control unit 11 of the sales activity analysis support system S causes the behavior analysis unit 104 to start loop processing for each piece of new behavior information 1402 that makes up the new behavior information list 1401. This starts loop processing for each piece of new behavior information 1402 that makes up the new behavior information list 1401. When the control unit 11 of the sales activity analysis support system S completes the processing in step S1105, it proceeds to step S1106.

[0102] In step S1106, the control unit 11 of the sales activity analysis support system S executes a process of recording the new behavior information 1402 in the behavior history database 106 using the behavior analysis unit 104. At this time, any behavior information 502 already stored in the behavior history database 106 may be acquired, and it may be determined whether the behavior information 502 indicates the same behavior as the new behavior information 1402. If one or more pieces of behavior information 502 that are determined to be identical already exist in the behavior history database 106, the new behavior information 1402 may be determined to be duplicate behavior information, and the recording process in step S1106 may be rejected. The above-mentioned rejection of the recording process may be based on another rule. For example, if manually registered behavior information 502 exists in the behavior history database 106 and it is determined that the new behavior information 1402 indicates the same behavior as the behavior information 502, the recording process of the new behavior information 1402 may be rejected. The process of determining whether the new behavior information 1402 and the behavior information 502 are identical may determine that the new behavior information 1402 and the behavior information 502 are identical when all information constituting the new behavior information 1402 (such as behavior time 1404 and subject 1405) matches all information constituting the behavior information 502 (such as behavior time 504 and subject 505). However, the determination of the above-mentioned consistency may be based on whether character strings match exactly, may be performed based on another rule designed in advance, may be performed using a statistical model, machine learning, deep learning model, or the like, or may be performed by combining these. For example, using a deep learning model that takes a character string as input and outputs a numerical sequence, the behavior content 1407 of the new behavior information 1402 and the behavior content 507 of the behavior information 502 may be converted into numerical sequences, the cosine similarity between the numerical sequences may be calculated, and if the cosine similarity exceeds a certain threshold, it may be determined that the behavior content 1407 of the new behavior information 1402 matches the behavior content 507 of the behavior information 502. Note that the control unit 11 of the sales activity analysis support system S can manually record, correct, and otherwise modify various information stored in the behavior history database 106, including the new behavior information 1402, by receiving input operations from a user via the input unit 51 and / or the communication unit 14.As a result, the new behavior information 1402 is recorded in the behavior history database 106. When the control unit 11 of the sales activity analysis support system S completes the process in step S1106, the process proceeds to step S1107.

[0103] In step S1107, the control unit 11 of the sales activity analysis support system S ends the loop process when the behavior analysis unit 104 performs the above process on all of the new behavior information 1402 that make up the new behavior information list 1401. When the control unit 11 of the sales activity analysis support system S ends the loop process, it ends the behavior analysis process shown in the flowchart of FIG.

[0104] (Sales behavior analysis processing) FIG. 14 is a flowchart showing an example of the flow of the sales behavior analysis process in the first embodiment (and the second embodiment).

[0105] The control unit 11 of the sales activity analysis support system S executes the above-mentioned sales behavior analysis process by the sales behavior analysis unit 105, which receives identification information that uniquely identifies a case and the calendar display month, and displays a calendar screen 1801 showing a summary of the sales representative's behavior based on the behavior information 1602 that has been stored in the behavior history database 106 and has the same case ID 1603 as the identification information. Details of this sales behavior analysis process will be explained using Figures 14 to 17.

[0106] In step S1401, the control unit 11 of the sales activity analysis support system S executes processing in which the sales behavior analysis unit 105 receives identification information that uniquely identifies the case and the display month 1806 on the calendar. However, the identification information and the display month 1806 may be input by a user checking the calendar screen 1801, or may be input by an external system. In this way, the identification information that uniquely identifies the case and the display month 1806 on the calendar are acquired. When the processing in step S1401 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1402.

[0107] In step S1402, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 acquires, from the action history database 106, an action information list 1601 that is configured from action information 1602 that has already been stored in the action history database 106 and has the same case ID 1603 as the identification information. In the first embodiment, the action information 1602 acquired in step S1402 is configured from the case ID 1603, action time 1604, subject 1605, subject position 1606, action content 1607, and recording date 1608. However, the action information 1602 may include items other than those described above. In this way, the action information list 1601 is acquired from the action history database 106. When the process in step S1402 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1403.

[0108] In step S1403, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to start loop processing for each piece of behavior information 1602 that constitutes the behavior information list 1601. This starts loop processing for each piece of behavior information 1602 that constitutes the behavior information list 1601. When the processing in step S1403 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1404.

[0109] In step S1404, the control unit 11 of the sales activity analysis support system S causes the sales behavior analysis unit 105 to execute a process of determining whether the subject's position 1606 in the behavioral information 1602 is a character string indicating a sales representative. For example, character strings such as "person in charge," "sales representative," and "person in charge" may be defined, and if any of these character strings is included in the subject's position 1606, the subject's position 1606 may be determined to be a character string indicating a sales representative. However, the above-mentioned determination process may be performed based on another rule, or may be performed using a statistical model, machine learning, deep learning model, or the like, or a combination thereof. For example, a machine learning model may be used that inputs the subject's position 1606 and outputs a true / false value indicating whether the subject is a sales representative. If it is determined in step S1404 that the subject's position 1606 is a character string indicating a sales representative (step S1404: YES), the process proceeds to step S1405. On the other hand, if it is determined in step S1404 that the subject position 1606 is not a character string indicating a sales representative (step S1404: NO), the process proceeds to step S1408, where the loop process ends.

[0110] In step S1405, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 normalizes the action time 1604 of the action information 1602 into an arbitrary format representing the year, month, and date, and adds the normalized value to the action information 1602 as a normalized action time 1610. In the first embodiment, it is assumed that the action time 1604 is a character string, and a process of normalizing it into a format that can identify the year, month, and date, such as "year-month-day" (e.g., 2022-06-12), is described. In the first embodiment, the action time 1604 is matched with a regular expression to identify each of the year, month, and date. For example, if the action time 1604 contains a character string such as "January 25, 2020," "August 6," "mid-July," or "around the 15th" using a regular expression such as "((1[0-9])|(2[0-9])|([30|1])|[1-9]) day," the year, month, and date can be detected. In the action time 1604 in the above example, part of the year, month, and date cannot be identified. The unidentifiable year, month, and date is filled in with an arbitrary tag indicating a missing value, and the identified year, month, and date is converted into a normalized action time 1610 of "year-month-day." For example, in the above-mentioned regular expression matching process, if the action time 1604 matches the character string "mid-July," it is converted into a normalized action time 1610 such as "<missing>-07-<missing>." During the above-mentioned process, the missing part of the year, month, and date is completed based on the expression of the character string matched against the regular expression. For example, if the character string is "middle of the month," the value "15" is interpolated. For example, if the character string is "end of the month," the values ​​"29," "30," and "31" are interpolated to match the month value of the normalized behavior period 1610. For the year of the normalized behavior period 1610 that remains missing after the above-mentioned process, the same year as the recording date 1608 of the behavior information 1602 is assigned. Note that in the above-mentioned process, if the date determined by interpolating the missing year, month, and day of the normalized behavior period 1610 with the year, month, and day of the recording date 1608 does not exist among the dates before the recording date 1608, the year assigned to the missing year of the normalized behavior period 1610 is the year immediately preceding the year of the recording date 1608. For the month of the normalized behavior period 1610 that remains missing after the above-mentioned process, the same month as the recording date 1608 of the behavior information 1602 is assigned.If the normalized behavior time 1610 still contains missing dates after the above-described process, the behavior information 1602 is deleted from the behavior information list 1601, and the process for the next behavior information 1602 is resumed from step S1403. However, the normalization process for the behavior time 1604 may be performed without performing the above-described deletion process. The normalization process in step S1405 may be performed using a different rule, a statistical model, a machine learning / deep learning model, or a combination of these methods. Furthermore, the normalization process in step S1405 may be performed by referring to something other than the behavior time 1604 or the recording date 1608 of the behavior information 1602. As a result, the behavior time 1604 of the behavior information 1602 is normalized to an arbitrary format representing the date, and the normalized value is added to the behavior information 1602 as the normalized behavior time 1610. When the control unit 11 of the sales activity analysis support system S completes the process in step S1405, the process proceeds to step S1406.

[0111] In step S1406, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 summarizes the action content 1607 of the action information 1602 to an arbitrary number of characters to generate summarized action content 1711 and adds the summarized action content 1711 to the action information 1602. Note that summarization refers to a process of converting the character string of the action content 1607 into a character string with fewer characters than the original character string. The summarization process may be performed based on pre-designed rules, or may be performed using a statistical model, machine learning, deep learning model, or the like, or a combination of these. However, the action content 1607 may be directly used as the summarized action content 1711 without summarization. As a result, the summarized action content 1711 is generated and added to the action information 1602. After completing the process in step S1406, the control unit 11 of the sales activity analysis support system S proceeds to step S1407.

[0112] In step S1407, the control unit 11 of the sales activity analysis support system S executes processing to add the behavior information 1602 to the display behavior information list 1701 as display behavior information 1702 using the sales behavior analysis unit 105. However, not all items of the behavior information 1602 need to be inherited to the display behavior information 1702. As a result, the behavior information 1602 is added to the display behavior information list 1701 as display behavior information 1702. When the processing in step S1407 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1408.

[0113] In step S1408, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to perform the above processing on all of the behavior information 1602 that constitutes the behavior information list 1601, and then ends the loop processing.

[0114] In step S1409, the control unit 11 of the sales activity analysis support system S executes processing to display the calendar screen 1801 using the display behavior information list 1701, using the sales behavior analysis unit 105, based on the calendar display month 1806 acquired in step S1401. However, in the first embodiment, the allowable reliability 1808 of the calendar screen 1801 is not displayed. When the calendar display starts on a Sunday, if the first day of the display month 1806 is a Sunday, the first day is set as the calendar display start date. If the first day of the display month 1806 is not a Sunday, the Sunday immediately preceding the first day is set as the calendar display start date. Also, if the last day of the display month 1806 is a Saturday, the last day is set as the calendar display end date. If the last day is not a Saturday, the Saturday immediately following the last day is set as the display end date. However, the calendar display does not have to start on a Sunday. If it starts on another day of the week, the display start date and display end date are calculated according to the starting day of the week. The number of weeks to be displayed is calculated by dividing (the display end date - the display start date + 1) by 7. A table with 7 columns and the number of weeks to be displayed + 1 rows is created as the calendar 1803. In the first row of the calendar 1803, the days of the week are entered in the order of the day of the week on which display starts in the calendar 1803. In the i+1 row of the calendar 1803, the days from (7 × (i-1)) days after the display start date to (7 × i-1) days after the display start date are entered in order, starting from the leftmost cell. Here, the aforementioned i represents an integer from 1 to the number of weeks to be displayed. The display behavior information 1702 having, in the normalization behavior period 1710, a date that corresponds to either the display end date, the display start date, or a date between the display end date and the display start date is selected as normalized behavior information. For each selected normalized behavior information (behavior information for display 1702), the summary behavior content 1711 of the normalized behavior information (behavior information for display 1702) is added to the cell where the same date as the normalized behavior time 1710 of the normalized behavior information (behavior information for display 1702) is entered. After the above-mentioned processing, the identification information that uniquely identifies the case accepted in step S1401 is set as a case ID 1802, and the case ID 1802, the calendar 1803, and the display month 1806 are arranged on the calendar screen 1801 according to Fig. 17 .For each piece of normalized behavior information (behavior information for display 1702) that has been added to the calendar 1803, a portion of the information contained in the normalized behavior information (behavior information for display 1702) is displayed on the calendar screen 1801 as individual details 1805. Alternatively, a separate database may be prepared to store the input document 101 received by the stored information summarizing unit 102, and the new behavior information 1402 generated by the behavior analysis unit 104 may include identification information that uniquely identifies the input document 101. Alternatively, the input document 101 may be included in the new behavior information 1402 generated by the behavior analysis unit 104, and the input document 101 that is the source of generation of the normalized behavior information (behavior information for display 1702) may be displayed on the calendar screen 1801 as a basis document 1807. However, the basis document 1807 does not necessarily have to be displayed. After the above processing is completed, the calendar screen 1801 is displayed. As a result, the calendar screen 1801 is displayed using the display behavior information list 1701 based on the display month 1806 of the calendar. Upon completing the processing in step S1409, the control unit 11 of the sales activity analysis support system S ends the sales behavior analysis processing shown in the flowchart of FIG.

[0115] According to the first embodiment, it is possible to comprehensively and automatically record diverse and detailed behavioral details that could not be obtained by conventional techniques based on the contents of the input document 101, and therefore it is possible to grasp more detailed and diverse behaviors of salespeople on the calendar screen 1801. In addition, since new behavioral information 1402 generated based on the input document 101 is accumulated, it is also possible to display the basis document 1807 on the calendar screen 1801, and the viewer of the calendar screen 1801 can grasp more detailed behavioral details of salespeople according to his or her own desires.

[0116] The sales activity analysis support system S in the first embodiment has been described above. [Example]

[0117] Next, the sales activity analysis support system S in the second embodiment will be described, focusing on the differences from the sales activity analysis support system S in the first embodiment.

[0118] <System configuration example> First, a configuration example of a sales activity analysis support system S in the second embodiment will be described with reference to FIG. 1 and FIGS. 18 and 19. FIG.

[0119] Fig. 1 is a diagram showing an example of the hardware configuration of an information processing device J constituting a sales activity analysis support system S in Example 2 (and Examples 1 and 3). In addition, Fig. 18 and Fig. 19 are both diagrams showing examples of functional blocks of the sales activity analysis support system S in Example 2.

[0120] (Example of the configuration of the sales activity analysis support system S) The sales activity analysis support system S in the second embodiment is a computer system that enables detailed and efficient review of the actions of salespeople in sales activities, thereby supporting the salespeople in analyzing their sales activities, and is realized by a computer device or server device similar to that in the first embodiment.

[0121] (Example of hardware configuration for the Sales Activity Analysis Support System S) That is, the hardware configuration of the information processing device J constituting the sales activity analysis support system S in the second embodiment is as illustrated in Fig. 1, and is the same as the hardware configuration of the information processing device J in the first embodiment. Therefore, a description of the example hardware configuration of the information processing device J in the second embodiment will be omitted.

[0122] (Example of functional blocks of the sales activity analysis support system S) Next, an example of various functional blocks included in the sales activity analysis support system S in the second embodiment will be described with reference to Figures 18 to 19. Note that each block described below does not represent a hardware configuration, but represents a functional block.

[0123] As illustrated in FIGS. 18 and 19, the control unit 11 of the sales activity analysis support system S in the second embodiment further includes a document type identification unit 108 as a functional block.

[0124] The document type identification unit 108 executes a process related to document type identification (hereinafter also referred to as a "document type identification process") (described in detail later).

[0125] That is, when the sales activity analysis support program (not shown) is executed by the processor 1, the functions of the document type identification unit 108 are realized and the document type identification process is performed.

[0126] Furthermore, in the sales activity analysis support system S in the second embodiment, the content of the process inference process executed by the process inference unit 103 (described in detail later in relation to FIG. 22) is different from the content of the process inference process in the first embodiment (described in detail above in relation to FIG. 9). Therefore, the difference in configuration between the sales activity analysis support system S in the second embodiment and the sales activity analysis support system S in the first embodiment is that the process inference unit 103 acquires various pieces of information stored in the action history database 106 and the case management database 107, as illustrated in FIG.

[0127] The other configurations are the same as those shown in FIGS. 2 and 3 as an example of blocks of various functions provided in the sales activity analysis support system S in the first embodiment, and therefore the description thereof will be omitted.

[0128] <Processing flow example> Next, the above-mentioned processes executed by the sales activity analysis support system S in the second embodiment will be described with reference to FIGS. 5 to 8, 11 to 17, and 20 to 26. FIG.

[0129] (Stored information summary processing) FIG. 20 is a flowchart illustrating an example of the flow of the stored information summarizing process according to the second embodiment.

[0130] When the control unit 11 of the sales activity analysis support system S receives the input document 302 and the case identification information 305 that uniquely identifies the sales case for which the input document 302 was created, the control unit 11 executes the stored information summarization process described above, in which the stored information summarization unit 102 retrieves case management information 401 related to the sales case from the case management database 107 and retrieves behavior information 502 related to the sales case that has been stored in the behavior history database 106, and generates summary data 701 based on the extracted behavior information 502 and the case management information 401. Details of this stored information summarization process will be described using Fig. 20 and Figs. 5 to 8.

[0131] 20 are the same as the processes shown as steps S401 to S409 in the flowchart of Fig. 4 in the explanation of the first embodiment. That is, the process executed in step S2001 of Fig. 20 is the same as the process executed in step S401 of Fig. 4, the process executed in step S2002 of Fig. 20 is the same as the process executed in step S402 of Fig. 4, the process executed in step S2003 of Fig. 20 is the same as the process executed in step S403 of Fig. 4, the process executed in step S2004 of Fig. 20 is the same as the process executed in step S404 of Fig. 4, and the process executed in step S2005 of Fig. 20 is the same as the process executed in step S401 of Fig. 4. 4, the process performed in step S405 of FIG. 20 is the same as the process performed in step S406 of FIG. 4, the process performed in step S2007 of FIG. 20 is the same as the process performed in step S407 of FIG. 4, the process performed in step S2008 of FIG. 20 is the same as the process performed in step S408 of FIG. 4, and the process performed in step S2009 of FIG. 20 is the same as the process performed in step S409 of FIG. 4.

[0132] In step S2010, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 selects one or more pieces of behavioral information 502 from the behavioral information list 501 acquired in step S2005 and acquires the selected pieces of behavioral information 502 as past behavioral information 707. In the second embodiment, one piece of behavioral information 502 is selected, the behavioral information 502 having the behavioral time 504 closest to the description date 303 of the input data 301 or the date on which the input data 301 was received. However, another piece of behavioral information 502 may be selected, or multiple pieces of behavioral information 502 may be aggregated to create the past behavioral information 707. When calculating the proximity between the behavioral time 504 and the date on which the input document 302 was received, the character string of the behavioral time 504 is normalized to a format that identifies the date, such as "year-month-day" (e.g., 2021-06-15), using the same process as step S1405 in FIG. 14 described in the first embodiment. Then, the difference in days between the normalized behavior time 504 and the receipt date of the input document 302 is calculated, and the closer the difference in days, the closer the receipt date of the behavior time 504 and the input document 302 is considered to be. The difference in days between the behavior time 504 and the receipt date of the input document 302 may be calculated based on another rule that inputs the behavior time 504 and the description date 303 and calculates the difference in days, without normalizing the behavior time 504. It may also be calculated using a statistical model, machine learning, deep learning model, or a combination of these. As a result, one or more pieces of behavior information 502 selected from the behavior information list 501 are acquired as past behavior information 707. Upon completing the process in step S2010, the control unit 11 of the sales activity analysis support system S proceeds to step S2011.

[0133] In step S2011, the control unit 11 of the sales activity analysis support system S executes a process in which the accumulated information summarizing unit 102 combines the summary text 702 generated in step S2009, the input document 302, the case identification information 305 of the input document, and the past behavior information 707 acquired in step S2010 to generate summary data 701. As a result, the summary text 702, the input document 302, the case identification information 305 of the input document, and the past behavior information 707 are combined to generate the summary data 701. Upon completing the process in step S2011, the control unit 11 of the sales activity analysis support system S ends the accumulated information summarizing process shown in the flowchart of FIG. 20.

[0134] (Document type identification process) The control unit 11 of the sales activity analysis support system S executes the document type identification process described above, in which the document type identification unit 108 receives the input document 302 received by the stored information summarizing unit 102 and generates document type information 2907 for the input document 302. Details of this document type identification process will be described with reference to FIGS. 5 and 21.

[0135] First, the control unit 11 of the sales activity analysis support system S executes a process of accepting input data 301, which is a pair of an input document 302 and case identification information 305, via the document type identification unit 108. However, the input data 301 does not necessarily have to include the writing date 303 or the case identification information 305. After the aforementioned process of accepting the input data 301, document type information 2907 is generated based on the input document 302. The document type information 2907 is data that stores information indicating the description format of a character string described in text 304 of the input document 302 and the type of content of the character string, and may hold any item related to the content of the character string in the text 304 as information. In the second embodiment, the document type information 2907 includes at least a document format 2908 indicating the description format of the text 304, a topic 2909 indicating a rough type of the content of the text 304, and an inference phase 2910 indicating the stage of sales activity at which the content of the text 304 is likely to be mentioned. It is also possible to provide a separate document type database that stores a finite number of types for each item that constitutes the document type information 2907, and to refer to this database when generating each of the above-mentioned items. In the explanation of the second embodiment, it is assumed that the document type database exists, but it is not necessary to provide the document type database. In this way, the input data 301 is acquired. When the above process is completed, the control unit 11 of the sales activity analysis support system S proceeds to the next step.

[0136] Next, the control unit 11 of the sales activity analysis support system S executes a process in which the document type identification unit 108 generates each item constituting the document type information 2907 from the text 304. In generating the document format 2908, one of the document formats 2908 stored in the document type database, namely, "email," "minutes," "system message," "chat log," and "any document," is selected based on the text 304 to be used as the document format 2908 for the text 304. In the second embodiment, the document format 2908 is selected by matching a regular expression. For example, if a regular expression indicating a format such as "{any character string} (San|Sama|Dear|Dearest) {any character string} {character string indicating an email signature}" matches the text 304, the document format 2908 for the text 304 is set to "email." By preparing unique regular expressions for each of "email," "minutes," "system message," and "chat log," it is possible to determine which document format 2908 the text 304 belongs to. If the text 304 does not match any of the regular expressions for "email," "minutes," "system message," or "chat log," the document format 2908 of the text 304 is set to "other." However, the process of generating the document format 2908 may be performed based on other rules, or may be performed using a statistical model, machine learning, deep learning model, or a combination thereof. For example, a content-generating deep learning model that inputs the character string of the text 304 and outputs a character string indicating the document format 2908 may be used. In the above case, the document type database is unnecessary. Similarly to the document format 2908, the generation of the topic 2909 and the inference phase 2910 may also be performed based on predetermined rules, or may be performed using a statistical model, machine learning, deep learning model, or a combination thereof. However, when generating the topic 2909 and the inference phase 2910, a list of predefined types of topic 2909 and inference phase 2910 may be obtained from the document type database. After the process of creating the document type information 2907 described above, the document type identification unit 108 outputs the document type information 2907. As a result, each item constituting the document type information 2907 is generated from the text 304.When the above process is completed, the control unit 11 of the sales activity analysis support system S ends the document type identification process.

[0137] (History inference processing) FIG. 22 is a flowchart illustrating an example of the flow of the process of inferring circumstances in the second embodiment.

[0138] The control unit 11 of the sales activity analysis support system S executes the process of inferring the process described above by the process inference unit 103, which receives the summary data 701 output by the stored information summarizing unit 102 and generates process information 2911 indicating the process up to the reception of the input data 301. Details of this process inferring the process will be described with reference to FIGS.

[0139] In step S2201, the control unit 11 of the sales activity analysis support system S causes the process estimation unit 103 to execute processing to receive the summary data 701 output by the stored information summarizing unit 102 and the document type information 2907 output by the document type identification unit 108. This acquires the summary data 701 and the document type information 2907. Upon completing the processing in step S2201, the control unit 11 of the sales activity analysis support system S proceeds to step S2202.

[0140] In step S2202, the control unit 11 of the sales activity analysis support system S executes a process of acquiring, by the process estimation unit 103, a case management information list 2401 composed of multiple pieces of case management information 2402 from the case management database 107. Here, the case management information 2403 is data that itemizes values ​​indicating characteristics of a sales case and includes at least a case ID 2403. The case management information 2403 acquired by the process estimation unit 103 in the second embodiment includes a case ID 2403 that uniquely identifies the case, a sales representative 2404 that indicates the company's sales representative for the case, a customer name 2405 that indicates the name of the customer targeted by the case, a customer representative 2406 that indicates the customer representative for the case, a product 2407 handled in the case, and an industry 2408 to which the customer belongs. However, the case management information 2402 may include information on other stakeholders or more detailed profile information such as the department to which the sales representative 403 and the customer representative 405 belong. In addition, the case management information 401 may include various other information describing the case, such as the case's project structure, type, customer industry, size, expected order amount, case creation date, expected order date, lead time, importance, etc. When obtaining the case management information 2402 in step S2202, only similar case management information 2402 may be obtained from the case management database 107 for the case management information 2402 specified by the case identification information 706. For example, the case similarity between the case management information 2402 specified by the case identification information 706 and each piece of case management information 2402 in the case management database 107 is calculated, and only case management information 2402 whose case similarity exceeds a certain threshold is obtained as the case management information list 2401. In the second embodiment, the case similarity is assumed to be a numerical value ranging from 0 to 1, but the case similarity may be any range. The calculation of the case similarity may be performed for two pieces of case management information 2402 based on any rule that takes any information constituting both pieces of information as input and outputs a numerical value ranging from 0 to 1, or may be performed using a statistical model, machine learning, deep learning model, or a combination of these. In this way, the case management information list 2401 is acquired from the case management database 107.When the control unit 11 of the sales activity analysis support system S completes the process in step S2202, the process proceeds to step S2203.

[0141] In step S2203, the control unit 11 of the sales activity analysis support system S executes processing to acquire a cross-case action information list 2501 made up of action information 2502 from the action history database 106 using the process estimation unit 103. At this time, the control unit 11 acquires, as the cross-case action information list 2501, action information 2502 having a case ID 2503 that matches any of the case IDs 2403 of all the case management information 2402 making up the case management information list 2401. As a result, the cross-case action information list 2501 is acquired from the action history database 106. When the processing in step S2203 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2204.

[0142] In step S2204, the control unit 11 of the sales activity analysis support system S executes a process of causing the process estimation unit 103 to create a list of the behavior information 2502 constituting the cross-case behavior information list 2501 acquired in step S2203, of which the past behavior similarity 2610 with the past behavior information 707 of the summary data 701 exceeds a certain threshold, and extracting the list of the behavior information 2502 as a similar past behavior information list 2602. The calculation of the past behavior similarity 2610 may be performed by measuring item-by-item similarity for the same item held by the past behavior information 707 and the behavior information 2502 and aggregating the item-by-item similarity to obtain the past behavior similarity 2610, or by directly comparing the entire data of the past behavior information 707 and the behavior information 2502 to obtain the past behavior similarity 2610. In the second embodiment, the past behavior similarity 2610 is assumed to be an arbitrary numerical value ranging from 0 to 1. However, the past behavior similarity 2610 may be any arbitrary range. For example, using a deep learning model that converts text into a numerical string, the behavior content 2507 of the behavior information 2502 and the behavior content of the past behavior information 707 may be converted into numerical strings, and the cosine similarity of the numerical strings may be calculated to obtain the past behavior similarity 2610. However, the calculation of the past behavior similarity 2610 may be performed based on a different rule, or may be performed using a statistical model, a machine learning model, a deep learning model, or a combination thereof. As a result, the list of the behavior information 2502 is extracted as the similar past behavior information list 2602. After completing the process at step S2204, the control unit 11 of the sales activity analysis support system S proceeds to step S2205.

[0143] In step S2205, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 determines whether or not there is one or more pieces of similar past information 2603 that constitute the similar past behavior information list 2602. If it is determined in step S2205 that there is one or more pieces of similar past information 2603 that constitute the similar past behavior information list 2602 (step S2205: YES), the process proceeds to step S2206, where a loop process is started for each piece of similar past information 2603 that constitutes the similar past behavior information list 2602. On the other hand, if it is determined in step S2205 that there is not one or more pieces of similar past information 2603 that constitute the similar past behavior information list 2602 (step S2205: NO), the process proceeds to step S2212.

[0144] In step S2206, as described above, the control unit 11 of the sales activity analysis support system S causes the process estimation unit 103 to start loop processing for each piece of similar past information 2603 that constitutes the similar past behavior information list 2602. This starts loop processing for each piece of similar past information 2603 that constitutes the similar past behavior information list 2602. When the control unit 11 of the sales activity analysis support system S completes the processing in step S2206, the control unit 11 proceeds to step S2207.

[0145] In step S2207, the control unit 11 of the sales activity analysis support system S executes a process of acquiring a similar current behavior information list 2611 using the process estimation unit 103. In the second embodiment, the text 705 of the summary data 701 is first summarized into one sentence. However, any length of the summarized character string may be specified. Furthermore, the summarization may be performed based on rules, using a statistical model, a machine learning model, a deep learning model, or a combination thereof. After acquiring the summarized text through the above-described summarization process, a list of behavior information 2502 in the cross-case behavior information list 2501 that has the same case ID 2503 as the case ID 2604 of the similar past information 2603, in which the current behavior similarity 2619 between the summary text and the action content 2507 of the behavior information 2502 exceeds a certain threshold, is acquired as the similar current behavior information list 2611. In the second embodiment, the current behavior similarity 2619 is assumed to be a numerical value ranging from 0 to 1, but the past behavior similarity 2610 may be in any range. For example, the current behavior similarity 2619 is calculated using a deep learning model that converts the summary text and the behavior content 2507 into a value indicating the current behavior similarity 2619 using as input. However, the current behavior similarity 2619 may be calculated based on rules, or may be calculated using a statistical model, machine learning, deep learning model, or the like, or a combination thereof. In this way, the similar current behavior information list 2611 is acquired. Upon completing the process in step S2207, the control unit 11 of the sales activity analysis support system S proceeds to step S2208.

[0146] In step S2208, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 determines whether or not there is one or more pieces of similar current behavior information 2612 constituting the similar current behavior information list 2611. If it is determined in step S2208 that there is one or more pieces of similar current behavior information 2612 constituting the similar current behavior information list 2611 (step S2208: YES), the process proceeds to step S2209. On the other hand, if it is determined in step S2208 that there is not one or more pieces of similar current behavior information 2612 constituting the similar current behavior information list 2611 (step S2208: NO), the process proceeds to step S2211, and the loop process is terminated.

[0147] In step S2209, the control unit 11 of the sales activity analysis support system S first executes processing to acquire an inferred behavior information list 2805 in order to generate a history information candidate 2802 using the history estimation unit 103. Among the behavior information 2502 having the same case ID 2503 as the case ID 2603 of the similar past information 2603 in the cross-case behavior information list 2501, a list of the behavior information 2502 having an action time 2504 later than the action time 2605 of the similar past information 2603 is acquired as the inferred behavior information list 2805. The action time 2504 later than the action time 2605 of the similar past information 2603 is determined by normalizing the character string indicating the time into a format that allows the date to be specified, such as "year-month-day," using the same processing as in step S1405 described in the first embodiment, and then comparing the context of the date. Thereafter, to narrow down the inferred behavior information 2806 in the inferred behavior information list 2805, first, the similar current behavior information 2612 having the earliest behavior time 2614 in the similar current behavior information list 2611 is acquired as the anchor-like current behavior information. Note that when selecting anchor-like current behavior information under the condition of the previous sentence, if there are multiple anchor-like current behavior information candidates, the similar current behavior information 2612 with the earliest recording date 2618 among the candidates is selected as the anchor-like current behavior information. If there are multiple anchor-like current behavior information candidates under the condition of the previous sentence, the similar current behavior information 2612 with the highest current behavior similarity 2619 among the candidates is selected as the anchor-like current behavior information. If there are multiple anchor-like current behavior information candidates under the condition of the previous sentence, the anchor-like current behavior information is randomly selected from the candidates. Note that this selection process may be based on another rule. From the inferred behavior information list 2805 , inferred behavior information 2806 having a behavior time 2808 earlier than the behavior time 2614 of the anchor-similar current behavior information is extracted, and this list is used as the final inferred behavior information list 2805 .When comparing two behavioral periods 2614 or the behavioral period 2614 and the behavioral period 2808 to determine which behavioral period is earlier, the character string indicating the period is normalized to a format that identifies the date, such as "year-month-day," using the same process as in step S1405 described in the first embodiment, and then the length of the dates is compared. The past behavior similarity 2610 of the similar past information 2603 is acquired as the similarity 2803 of the past similar behavior. The current behavior similarity 2619 of the anchor similar current behavior information is acquired as the similarity 2804 of the current similar behavior. The inferred behavior information list 2805, the similarity 2803 of the past similar behavior, and the similarity 2804 of the current similar behavior are combined to create the history information candidate 2802. Thus, to generate the history information candidate 2802, the inferred behavior information list 2805 is first acquired. After completing the process in step S2209, the control unit 11 of the sales activity analysis support system S proceeds to step S2210.

[0148] In step S2210, the control unit 11 of the sales activity analysis support system S causes the process estimation unit 103 to execute processing to add the process information candidate 2802 to the process information candidate list 2801. As a result, the process information candidate 2802 is added to the process information candidate list 2801. When the process in step S2210 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2211.

[0149] In step S2211, the control unit 11 of the sales activity analysis support system S ends the loop processing by the process inference unit 103, and proceeds to step S2212. Note that, depending on the result of the conditional branch in step S2208, step S2210 may not be executed, and therefore the process information candidate list 2801 may contain zero process information candidates 2802.

[0150] In step S2212, the control unit 11 of the sales activity analysis support system S executes a process of selecting, by the process estimation unit 103, process information 2911 from the process information candidate list 2801 based on the inferred behavior information list 2805, the similarity 2803 of past similar behavior, and the similarity 2804 of current similar behavior. In the second embodiment, the process information candidate 2802 having the largest total similarity obtained by adding up the similarity 2803 of past similar behavior and the similarity 2804 of current similar behavior is selected as the process information 2911. Note that, when the case similarity is calculated in step S2202, the case similarity of the case management information 2402 having the same case ID 2403 as the case ID 2807 of the inferred behavior information 2806 may be added to the total similarity (according to step S2209, the case ID 2807 of all the inferred behavior information 2806 constituting the inferred behavior information list 2805 of each process information candidate 2802 is the same). When there are multiple history information candidates 2802 with the same total similarity, the history information candidate 2802 with the smallest number of inferred behavior information 2806 constituting the inferred behavior information list 2805 is selected as the history information 2911. When selecting the history information 2911 under the condition of the previous sentence, if there are multiple candidates for the history information 2911, the candidate with the latest recording date 2812 of the inferred behavior information 2806 constituting the inferred behavior information list 2805 of the candidate that is closest to the description date 704 of the summary data 701 is selected as the history information 2911. When selecting the history information 2911 under the condition of the previous sentence, if there are multiple candidates for the history information 2911, the history information 2911 is selected at random from among the candidates. If there are zero history information candidates 2802 in the history information candidate list 2801, the history information 2911 is generated such that the value of the similarity of past similar behavior 2912 is zero, the value of the similarity of current similar behavior 2913 is zero, and the inferred behavior information list 2914 has only zero inferred behavior information 2915. As a result, the history information 2911 is selected from the history information candidate list 2801 based on the inferred behavior information list 2805, the similarity of past similar behavior 2803, and the similarity of current similar behavior 2804. Upon completing the process in step S2212, the control unit 11 of the sales activity analysis support system S proceeds to step S2213.

[0151] In step S2213, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 combines the summary text 702 in the summary data 701, the input document 703, the case identification information 706, the document type information 2907, and the process information 2911 to create context data 2901. As a result, the summary text 702 in the summary data 701, the input document 703, the case identification information 706, the document type information 2907, and the process information 2911 are combined to create the context data 2901. When the process in step S2213 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2214.

[0152] In step S2214, the control unit 11 of the sales activity analysis support system S executes a process in which the process inference unit 103 outputs the context data 2901 created in step S2213. This creates the context data 2901. Upon completing the process in step S2214, the control unit 11 of the sales activity analysis support system S ends the process inference process shown in the flowchart of FIG.

[0153] (Behavioral analysis processing) As described above, FIG. 11 is a flowchart showing an example of the flow of the behavior analysis process in the second embodiment (and the first and third embodiments).

[0154] The control unit 11 of the sales activity analysis support system S executes the above-mentioned behavior analysis process by the behavior analysis unit 104, which receives the context data 2901 output in step S2214 of Fig. 22, generates zero or more pieces of behavior information as new behavior information 1402 based on the contents of the context data 2901, and records the generated new behavior information 1402 in the behavior history database 106. Details of this behavior analysis process will be described with reference to Figs. 11 to 13 and 21.

[0155] In step S1101, the control unit 11 of the sales activity analysis support system S executes processing to receive the context data 2901 output by the behavior analysis unit 104 in step S2214 of Fig. 22. This acquires the context data 2901. When the processing in step S1101 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1102.

[0156] In step S1102, the control unit 11 of the sales activity analysis support system S executes the process by the behavior analysis unit 104 to generate an analysis instruction text consisting of a character string of one or more characters, using the behavior analysis unit 104. However, in the second embodiment, the analysis instruction text is dynamically determined based on the context data 2901. In the second embodiment, the summary text 2902, the date of writing 2904, the text 2905, and the document type information 2907 are first applied to a template that reads "{the value of the summary text 2902} The following is the {the value of the document format 2908 of the document type information 2907} written in the {the value of the writing date 2904}. {the value of the text 2905} Note that, by the time the above {the value of the document format 2908 of the document type information 2907} is written," to generate the analysis instruction text. Next, starting from the first inferred behavior information 2915 in the inferred behavior information list 2914, an append string is created using the subject's position 2919 and the action content 2920 of the inferred behavior information 2915, which reads "{the value of the action content 2920} according to {the value of the subject's position 2919}," and the append string is added to the analysis instruction text. After performing the append process described above for all inferred behavior information 2915 in the inferred behavior information list 2914, the following string is added to the analysis instruction text: "The following actions may have been taken in advance. Based on the above situation, please infer all possible actions of each person from the contents of the above document, and list who (subject), to whom (object), and what they did for each action. In addition, please also tell us the subject's position in sales activities (sales representative, customer, third party, etc.). For each action, please list them in dictionary format with the subject name, subject's position, object name, action time, and action content as keys." Through the above process, an analysis instruction text is created. As a result, the analysis instruction text is created. Upon completion of the process in step S1102, the control unit 11 of the sales activity analysis support system S proceeds to step S1103.

[0157] In step S1103, the control unit 11 of the sales activity analysis support system S executes a process in which the behavior analysis unit 104 creates text data, which is a character string, using the context data 2901 acquired in step S1101, and generates a new behavior information list 1401 consisting of zero or more new behavior information 1402 based on the text data. The new behavior information 1402 may include items not acquired by the accumulated information summarizing unit 102. For example, in the second embodiment, the new behavior information 1402 includes a case ID 1403, a behavior time 1404, an agent 1405, an agent's position 1406, a behavior content 1407, a recording date 1408, and a reliability 1409. However, the new behavior information 1402 may include items other than those described above. The reliability 1409 is a measure of the credibility of the content indicated by the new behavior information 1402. In the process of step S1103 in the second embodiment, the analysis instruction text created in step S1102 is provided as input to a content-generating deep learning model. The output of the content-generating deep learning model is obtained as analysis result text 1302. However, the model used to generate the analysis result text 1302 does not have to be based on a deep learning model, and may be based on a statistical model or a machine learning model. Once the analysis result text 1302 is obtained from the content-generating deep learning model, the analysis result text 1302 is analyzed using regular expression-based string matching and converted into zero or more pieces of new behavior information 1402. For example, by writing a regular expression that detects a string pattern that starts with "{" and ends with "}" and has one or more patterns of "'character string': 'character string'", the section in the analysis result text 1302 in which the content of each piece of new behavior information 1402 is described can be identified. In each section of the aforementioned behavior information, a portion matching "character string":"character string" is detected, and the character string enclosed in """ is obtained, thereby obtaining the behavior time 1404, subject 1405, subject position 1406, and behavior content 1407 of the new behavior information 1402. The case identification information 2906 of the context data 2901 is added as the case ID 1403 to all the obtained new behavior information 1402. In addition, the entry date 2904 of the context data 2901 is added as the recording date 1408 to all the obtained new behavior information 1402.However, the date of writing 2904 may be the date on which the accumulated information summarizing unit 102 received the input document 1003. Furthermore, the reliability 1409 of the new behavior information 1402 is calculated based on the character string of the section in the analysis result text 1302 where the content of the new behavior information 1402 is described. In the second embodiment, the likelihood of the character string of the section in the deep learning model used to generate the analysis result text 1302 is calculated and acquired as the reliability 1409. In the second embodiment, the likelihood is assumed to be an arbitrary value ranging from 0 to 1. However, the calculation may be performed using any rule, statistical model, machine learning, or deep learning model that receives at least the character string of the section as input and calculates the reliability 1409, or a combination of these. For example, a deep learning model that receives two character strings, the character string of the section and the text 2905 of the context data 2901, and outputs a numerical value ranging from 0 to 1 may be used, and the output of the deep learning model may be acquired as the reliability 1409. The new behavior information 1402 created through the above process is arranged to create a new behavior information list 1401. The reliability 1409 of each piece of new behavior information 1402 constituting the new behavior information list 1401 may be readjusted using the history information 2911. For example, if there is inferred behavior information 2915 whose similarity between the new behavior information 1402 and the inferred behavior information 2915 exceeds a certain threshold, the reliability 1409 of the new behavior information 1402 may be replaced with the average value of the reliability 1409, the similarity 2912 of past similar behavior, and the similarity 2913 of current similar behavior. In the second embodiment, the similarity between the new behavior information 1402 and the inferred behavior information 2915 is assumed to be a numerical value ranging from 0 to 1. The similarity may be calculated based on rules, using a statistical model, a machine learning model, a deep learning model, or a combination thereof.For example, a deep learning model that takes two character strings as input and calculates the similarity between them may be used to calculate the similarity between the subject position 1406 of the new behavior information 1402 and the subject position 2919 of the inferred behavior information 2915 as the position similarity. If the position similarity is less than 0.99, the similarity between the new behavior information 1402 and the inferred behavior information 2915 may be set to 0. If the similarity is 0.99 or greater, the similarity output by the deep learning model when the action content 1407 of the new behavior information 1402 and the action content 2920 of the inferred behavior information 2915 are input may be set as the similarity between the new behavior information 1402 and the inferred behavior information 2915. In this way, the new behavior information list 1401 is generated based on the text data created using the context data 2901. After completing the process at step S1103, the control unit 11 of the sales activity analysis support system S proceeds to step S1104.

[0158] In step S1104, the control unit 11 of the sales activity analysis support system S, as in the first embodiment, executes a process in which the behavior analysis unit 104 determines whether the new behavior information list 1401 acquired in step S1103 is composed of one or more pieces of new behavior information 1402. If it is determined in step S1104 that the new behavior information list 1401 is composed of one or more pieces of new behavior information 1402 (step S1104: YES), the process proceeds to step S1105. On the other hand, if it is determined in step S1104 that the new behavior information list 1401 is not composed of one or more pieces of new behavior information 1402 (step S1104: NO), the behavior analysis process shown in the flowchart of FIG. 11 is terminated.

[0159] In step S1105, the control unit 11 of the sales activity analysis support system S, as in the first embodiment, causes the behavior analysis unit 104 to start loop processing for each piece of new behavior information 1402 that constitutes the new behavior information list 1401. This starts loop processing for each piece of new behavior information 1402 that constitutes the new behavior information list 1401. When the processing in step S1105 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1106.

[0160] In step S1106, the control unit 11 of the sales activity analysis support system S executes a process of recording the new behavior information 1402 in the behavior history database 106 using the behavior analysis unit 104, as in the case of the first embodiment. Note that also in the second embodiment, various information stored in the behavior history database 106, including the new behavior information 1402, can be manually recorded, corrected, or the like as needed by the control unit 11 of the sales activity analysis support system S accepting input operations from the user via the input unit 51 and / or the communication unit 14, as in the first embodiment. As a result, the new behavior information 1402 is recorded in the behavior history database 106. Upon completing the process in step S1106, the control unit 11 of the sales activity analysis support system S proceeds to step S1107.

[0161] In step S1107, the control unit 11 of the sales activity analysis support system S ends the loop process when the behavior analysis unit 104 performs the above process on all of the new behavior information 1402 that make up the new behavior information list 1401, as in the case of the first embodiment. When the control unit 11 of the sales activity analysis support system S ends the loop process, it ends the behavior analysis process shown in the flowchart of FIG.

[0162] (Sales behavior analysis processing) FIG. 14 is a flowchart showing an example of the flow of the sales behavior analysis process in the second embodiment (and the first embodiment).

[0163] The control unit 11 of the sales activity analysis support system S executes the above-mentioned sales behavior analysis process by the sales behavior analysis unit 105, which receives identification information that uniquely identifies a case, the month to be displayed on the calendar, and the allowable reliability for adjusting the behavior information to be displayed, and displays a calendar screen 1801 showing a brief behavior history of the sales representative based on behavior information 1602 that has been stored in the behavior history database 106 and has the same case ID 1603 as the identification information. Details of this sales behavior analysis process will be explained using Figures 14 to 17.

[0164] In step S1401, the control unit 11 of the sales activity analysis support system S executes processing in which the sales behavior analysis unit 105 receives identification information that uniquely identifies an instance, the calendar display month 1806, and the acceptable reliability 1808. However, the identification information, the calendar display month 1806, and the acceptable reliability 1808 may be input by a user checking the calendar screen 1801, or may be input by an external system. In this way, the identification information that uniquely identifies an instance, the calendar display month 1806, and the acceptable reliability 1808 are acquired. When the processing in step S1401 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1402.

[0165] In step S1402, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 acquires, from the action history database 106, an action information list 1601 that is configured from action information 1602 that has already been stored in the action history database 106 and has the same case ID 1603 as the identification information. In the second embodiment, the action information 1602 acquired in step S1402 is configured from the case ID 1603, action time 1604, subject 1605, subject position 1606, action content 1607, recording date 1608, and reliability 1609. However, the action information 1602 may include items other than those described above. In this way, the action information list 1601 is acquired from the action history database 106. When the process in step S1402 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1403.

[0166] In step S1403, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to start loop processing for each piece of behavior information 1602 that constitutes the behavior information list 1601. This starts loop processing for each piece of behavior information 1602 that constitutes the behavior information list 1601. When the processing in step S1403 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1404.

[0167] In step S1404, the control unit 11 of the sales activity analysis support system S causes the sales behavior analysis unit 105 to execute a process of determining whether the subject position 1606 of the behavior information 1602 is a character string indicating a sales representative. This determination process may be performed based on rules, or may be performed using a statistical model, machine learning, deep learning model, or the like, or a combination thereof. For example, a deep learning model may be used that inputs the subject position 1606 and the character string "sales representative" and outputs the similarity between the two character strings, and the subject position 1606 whose similarity exceeds a certain threshold may be determined to be a character string indicating a sales representative. If it is determined in step S1404 that the subject position 1606 is a character string indicating a sales representative (step S1404: YES), the process proceeds to step S1405. On the other hand, if it is determined in step S1404 that the subject position 1606 is not a character string indicating a sales representative (step S1404: NO), the process proceeds to step S1408, where the loop process ends.

[0168] In step S1405, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 normalizes the action time 1604 of the action information 1602 to an arbitrary format representing the year, month, and date, and adds the normalized value to the action information 1602 as the normalized action time 1610. The method for implementing this normalization process is the same as in the first embodiment. As a result, the action time 1604 of the action information 1602 is normalized to an arbitrary format representing the year, month, and date, and the normalized value is added to the action information 1602 as the normalized action time 1610. When the process in step S1405 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1406.

[0169] In step S1406, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 summarizes the action content 1607 of the action information 1602 to an arbitrary number of characters to generate summarized action content 1711 and adds the summarized action content 1711 to the action information 1602. Note that summarization refers to a process of converting the character string of the action content 1607 into a character string with fewer characters than the original character string. The summarization process may be performed based on pre-designed rules, or may be performed using a statistical model, machine learning, deep learning model, or the like, or a combination of these. However, the action content 1607 may be directly used as the summarized action content 1711 without summarization. As a result, the summarized action content 1711 is generated and added to the action information 1602. After completing the process in step S1406, the control unit 11 of the sales activity analysis support system S proceeds to step S1407.

[0170] In step S1407, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 extracts, from the behavior information 1602 constituting the behavior information list 1601, the behavior information 1602 having a reliability 1609 equal to or greater than the allowable reliability 1808 as the display behavior information 1702, and creates a display behavior information list 1701 made up of the display behavior information 1702. However, it is not necessary to carry over all the items of the behavior information 1602 to the display behavior information 1702. As a result, from the behavior information 1602 constituting the behavior information list 1601, the behavior information 1602 having a reliability 1609 equal to or greater than the allowable reliability 1808 is extracted as the display behavior information 1702, and the display behavior information list 1701 made up of the display behavior information 1702 is created. When the process in step S1407 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S1408.

[0171] In step S1408, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to perform the above processing on all of the behavior information 1602 that constitutes the behavior information list 1601, and then ends the loop processing.

[0172] In step S1409, the control unit 11 of the sales activity analysis support system S executes a process of generating a calendar screen 1801 based on the behavior information list for display 1701 using the sales behavior analysis unit 105, and displaying the calendar screen 1801, as in the first embodiment. However, in the second embodiment, the acceptable reliability 1808 acquired in step S1401 is displayed in the upper right corner of the calendar screen 1801. Note that the display position may be set to another location. As a result, the calendar screen 1801 generated based on the behavior information list for display 1701 is displayed. When the process in step S1409 is completed, the control unit 11 of the sales activity analysis support system S ends the sales behavior analysis process shown in the flowchart of FIG. 14.

[0173] According to the second embodiment, in addition to the effects of the first embodiment, the user can select whether to prioritize the reliability or the comprehensiveness of the displayed information by adjusting the allowable reliability 1808. That is, the accuracy of the analysis is improved.

[0174] The sales activity analysis support system S in the second embodiment has been described above. [Example]

[0175] Next, the sales activity analysis support system S in the third embodiment will be described, focusing on the differences from the sales activity analysis support system S in the first embodiment and / or the second embodiment.

[0176] <System configuration example> First, a configuration example of a sales activity analysis support system S in the third embodiment will be described with reference to FIGS. 1, 3, and 18-19.

[0177] Fig. 1 is a diagram showing an example of the hardware configuration of an information processing device J constituting the sales activity analysis support system S in the third embodiment (and the first and second embodiments). Fig. 18 is a diagram showing an example of the functional blocks of the sales activity analysis support system S in the third embodiment (and the second embodiment).

[0178] (Example of the configuration of the sales activity analysis support system S) The sales activity analysis support system S in Example 3 is a computer system that can analyze the behavior of salespeople and stakeholders in sales activities and identify factors that lead to orders, thereby supporting the analysis of sales activities, and is realized by a computer device or server device similar to those in Examples 1 and 2.

[0179] (Example of hardware configuration for the Sales Activity Analysis Support System S) That is, the hardware configuration of the information processing device J constituting the sales activity analysis support system S in the third embodiment is as exemplified in Fig. 1, and is the same as the hardware configuration of the information processing device J in the first and second embodiments. Therefore, a description of the hardware configuration example of the information processing device J in the third embodiment will be omitted.

[0180] (Example of functional blocks of the sales activity analysis support system S) Next, an example of various functional blocks included in the sales activity analysis support system S in the third embodiment will be described with reference to Fig. 3 and Figs. 18 to 19. Note that each block described below does not represent a hardware configuration but represents a functional block.

[0181] The sales activity analysis support system S in the third embodiment has a configuration in which a document type identification unit 108 shown in FIGS. 18 and 19 is further added to the configuration shown in FIG.

[0182] The other configurations are the same as those shown in FIGS. 2 and 3 as an example of blocks of various functions provided in the sales activity analysis support system S in the first embodiment, and therefore the description thereof will be omitted.

[0183] <Processing flow example> Next, the above-mentioned processes executed by the sales activity analysis support system S in the third embodiment will be described with reference to FIGS. 4 to 5, 7 to 13, 15, 21, and 27 to 30. FIG.

[0184] (Stored information summary processing) FIG. 4 is a flowchart showing an example of the flow of the stored information summarizing process in the third embodiment (and the first embodiment).

[0185] The control unit 11 of the sales activity analysis support system S, upon receiving an input document 302 and case identification information 305 that uniquely identifies the sales case for which the input document 302 was created, executes the stored information summarization process described above, in which the stored information summarization unit 102 retrieves case management information 401 related to the sales case from the case management database 107 and retrieves behavior information 502 related to the sales case that has already been stored in the behavior history database 106, and generates summary data 701 based on the extracted behavior information 502 and case management information 401. The details of the stored information summarization process in the third embodiment are the same as those in the first embodiment. That is, the stored information summarization unit 102 outputs summary data 701 shown in FIG. 8 based on the input document and the case identification information.

[0186] (Document type identification process) The control unit 11 of the sales activity analysis support system S receives the input document 302 received by the accumulated information summarizing unit 102 as the document type identification process described above, and executes a process of generating document type information 2907 for the input document 302 using the document type identification unit 108. The content of the document type identification process in the third embodiment is the same as that in the second embodiment. That is, the document type identification unit 108 outputs the document type information 2907 in FIG. 21 based on the input document 302.

[0187] (History inference processing) FIG. 9 is a flowchart showing an example of the flow of the process of inferring circumstances in the third embodiment (and the first embodiment).

[0188] The control unit 11 of the sales activity analysis support system S executes the process of inferring the process described above by the process inference unit 103, which receives the summary data 701 output by the stored information summarizing unit 102 and the document type information 2907 output by the document type identifying unit 108, and generates process information 1009 indicating the process up to the reception of the input data 301. Details of this process inferring the process in the third embodiment will be described with reference to Figs. 5, 8 to 10, and 21.

[0189] In step S901, the control unit 11 of the sales activity analysis support system S causes the process estimation unit 103 to execute processing to receive the summary data 701 and document type information 2907 output by the stored information summarizing unit 102. This acquires the summary data 701 and the document type information 2907. Upon completing the processing in step S901, the control unit 11 of the sales activity analysis support system S proceeds to step S902.

[0190] In step S902, the control unit 11 of the sales activity analysis support system S executes a process of generating the inference instruction text 1008 using the input document 703 and the document type information 2907 of the summary data 701 by the process inference unit 103. In the third embodiment, the inference instruction text 1008 is generated by applying the description date 704 and the text 705 of the input document 703 and the document format 2908 and topic 2909 of the document type information 2907 to a template that reads, "The following is about this business case, and is registered in {the value of the description date 704}, and the topic is {the topic 2909}. How was this {the value of the document format 2908} written? Please infer the process. {the value of the document format 2908} registered in {the value of the description date 704}: {the value of the text 705}." However, a different template may be used. As a result, the inference instruction text 1008 is created using the input document 703 of the summary data 701 and the document type information 2907. When the control unit 11 of the sales activity analysis support system S completes the process in step S902, the process proceeds to step S903.

[0191] In step S903, the control unit 11 of the sales activity analysis support system S causes the process inference unit 103 to perform inference processing using the summary data 701 and the inference instruction text 1008, and executes processing to generate process information 1009. The content of the processing in step S903 in the third embodiment is the same as that in the first embodiment. As a result, the inference processing is performed using the summary data 701 and the inference instruction text 1008, and process information 1009 is generated. When the processing in step S903 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S904.

[0192] In step S904, the control unit 11 of the sales activity analysis support system S executes a process in which the process estimation unit 103 combines the input document 302, the case identification information 305, the summary data 701 acquired in step S901, the inference instruction text 1008 created in step S902, and the process information 1009 generated in step S903 to create context data 1001. As a result, the input document 302, the case identification information 305, the summary data 701, the inference instruction text 1008, and the process information 1009 are combined to create the context data 1001. When the process in step S904 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S905.

[0193] In step S905, the control unit 11 of the sales activity analysis support system S executes a process of outputting the context data 1001 created in step S904 by the process inference unit 103, as in the first embodiment. This outputs the context data 1001. When the process in step S905 is completed, the control unit 11 of the sales activity analysis support system S ends the process inference process shown in the flowchart of FIG.

[0194] (Behavioral analysis processing) FIG. 11 is a flowchart showing an example of the flow of the behavior analysis process in the third embodiment (and the first embodiment).

[0195] As the above-mentioned behavior analysis process, the control unit 11 of the sales activity analysis support system S executes a process in which the behavior analysis unit 104 receives the context data 1001 output by the process estimation unit 103, generates zero or more pieces of behavior information as new behavior information 1402 based on the contents of the context data 1001, and records the generated new behavior information 1402 in the behavior history database 106. Note that, in the third embodiment, various pieces of information accumulated in the behavior history database 106, including the new behavior information 1402, can be manually recorded, corrected, or the like as appropriate by the control unit 11 of the sales activity analysis support system S receiving input operations from the user via the input unit 51 and / or the communication unit 14, as in the first embodiment. The contents of the behavior analysis process in the third embodiment are the same as those in the first embodiment.

[0196] (Sales behavior analysis processing) FIG. 27 is a flowchart illustrating an example of the flow of the sales behavior analysis process according to the third embodiment.

[0197] The control unit 11 of the sales activity analysis support system S executes the above-mentioned sales behavior analysis process by the sales behavior analysis unit 105, which receives specification information 3802 indicating the analysis settings, acquires behavior information 1602 corresponding to the specified behavior content 3804 by the specifying entity 3803 from the behavior history database 106, calculates the winning rate 3807 of the sales case for each category name 3806 of the next behavior performed within a time range 3805 for the behavior information 1602, and displays the winning rate 3807 for each behavior category on the analysis screen 3801. Details of this sales behavior analysis process will be described using Fig. 15 and Figs. 27 to 30.

[0198] In step S2701, the control unit 11 of the sales activity analysis support system S executes processing for receiving, via the sales behavior analysis unit 105, specification information 3802 indicating the analysis settings. The specification information 3802 is composed of a specification subject 3803, a specified behavior content 3804, and a time range 3805 for specifying the behavior to be analyzed. However, the specification information 3802 may be input by a user operating the analysis screen, or may be input by an external system. In this way, the specification information 3802 indicating the analysis settings is acquired. When the processing in step S2701 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2702.

[0199] In step S2702, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 acquires, from the action history database 106, an action information list 1601 that is already stored in the action history database 106 and that is configured from the action information 1602 that corresponds to the specified information 3802. In the third embodiment, the action information 1602 acquired in step S2702 is configured from a case ID 1603, an action time 1604, an agent 1605, an agent's position 1606, an action content 1607, and a recording date 1608. However, the action information 1602 may include items other than those described above. In step S2702 of the third embodiment, if it is determined that the subject position 1606 of the behavior information 1602 matches the designated subject 3803 and it is also determined that the behavior content 1607 of the behavior information 1602 matches the designated behavior content 3804, it is determined that the behavior information 1602 corresponds to the designated information 3802. The determination of whether the subject position 1606 matches the designated subject 3803 may be performed using any rule, a statistical model, a machine learning / deep learning model, or the like, or a combination of these means. In the third embodiment, if the character string of the designated subject 3803 is included in the character string of the subject position 1606, it is determined that the subject position 1606 matches the designated subject 3803. The determination of whether the action content 1607 of the action information 1602 matches the specified action content 3804 may be performed using any rule, a statistical model, a machine learning / deep learning model, or a combination of these means. In the third embodiment, a deep learning model is used to determine whether the character string of the action content 1607 matches the character string of the specified action content 3804 as input. As a result, the action information list 1601 is acquired from the action history database 106. Upon completing the process at step S2702, the control unit 11 of the sales activity analysis support system S proceeds to step S2703.

[0200] In step S2703, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to start loop processing for each piece of behavior information 1602 constituting the behavior information list 1601 acquired in step S2702. This starts loop processing for each piece of behavior information 1602 constituting the behavior information list 1601. When the control unit 11 of the sales activity analysis support system S completes the processing in step S2703, it proceeds to step S2704.

[0201] In step S2704, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 normalizes the action time 1604 of the action information 1602 to an arbitrary format representing the year, month, and date, and adds the normalized value to the action information 1602 as the normalized action time 1610. In the third embodiment, the action time 1604 is normalized to a format that identifies the year, month, and date, such as "year-month-day" (e.g., 2022-06-12), in the same manner as the process described in step S1405 of FIG. 14 in the first embodiment. As a result, the action time 1604 of the action information 1602 is normalized to an arbitrary format representing the year, month, and date, and the normalized value is added to the action information 1602 as the normalized action time 1610. When the process in step S2704 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2705.

[0202] In step S2705, the control unit 11 of the sales activity analysis support system S executes a process in which the sales behavior analysis unit 105 acquires, from the behavior history database 106, all of the behavior information that has already been accumulated in the behavior history database 106 and has the same case ID 3603 as the case ID 1603 of the behavior information 1602, as candidate analysis behavior information 3602, and creates a candidate analysis behavior information list 3601 made up of the candidate analysis behavior information 3602. However, at the time of step S2705, the candidate analysis behavior information 3602 does not include the normalized behavior period 3609. As a result, all of the behavior information that has already been accumulated in the behavior history database 106 and has the same case ID 3603 as the case ID 1603 of the behavior information 1602 is acquired from the behavior history database 106 as candidate analysis behavior information 3602, and a candidate analysis behavior information list 3601 made up of the candidate analysis behavior information 3602 is created. When the control unit 11 of the sales activity analysis support system S completes the process in step S2705, the process proceeds to step S2706.

[0203] In step S2706, the control unit 11 of the sales activity analysis support system S causes the sales behavior analysis unit 105 to start loop processing for each piece of analysis candidate behavior information 3602 that makes up the analysis candidate behavior information list 3601. This starts loop processing for each piece of analysis candidate behavior information 3602 that makes up the analysis candidate behavior information list 3601. When the control unit 11 of the sales activity analysis support system S completes the processing in step S2706, it proceeds to step S2707.

[0204] In step S2707, the control unit 11 of the sales activity analysis support system S executes a process in which the sales activity analysis unit 105 normalizes the behavior time 3604 of the candidate analysis behavior information 3602 to an arbitrary format representing the year, month, and date, converts it into a normalized behavior time 3609, and adds it to the candidate analysis behavior information 3602. In the third embodiment, normalization is performed to a format that allows the year, month, and date to be identified, such as "year-month-day" (e.g., 2022-06-12), using the same process as described in step S1405 of FIG. 14 in the first embodiment. As a result, the behavior time 3604 of the candidate analysis behavior information 3602 is normalized to an arbitrary format representing the year, month, and date, converts it into a normalized behavior time 3609, and adds it to the candidate analysis behavior information 3602. When the process in step S2707 is completed, the control unit 11 of the sales activity analysis support system S proceeds to step S2708.

[0205] In step S2708, the control unit 11 of the sales activity analysis support system S causes the sales behavior analysis unit 105 to perform the processing of step S2707 on all the analysis candidate behavior information 3602 that make up the analysis candidate behavior information list 3601, and then ends the loop processing.

[0206] In step S2709, the control unit 11 of the sales activity analysis support system S executes a process in which the sales behavior analysis unit 105 selects zero or more pieces of analysis candidate behavior information 3602 from the analysis candidate behavior information list 3601 as behavior information for analysis 3702, and adds the pieces to the behavior information for analysis list 3701. In the third embodiment, in step S2709, first, the date difference is calculated for each piece of analysis candidate behavior information 3602 in the analysis candidate behavior information list 3601 using the following formula:

[0207] Date difference = normalized behavior period 3609 of the analysis candidate behavior information 3602 - normalized behavior period 1610 of the behavior information 1602

[0208] The analysis candidate behavior information 3602 for which the date difference is greater than 0 and less than or equal to the value indicated by the time range 3805 of the specification information 3802 is selected, and of the candidate behavior information 3602 for analysis selected by the above-mentioned process, the analysis candidate behavior information 3602 having the earliest date in the normalized behavior period 3609 is added to the analysis candidate behavior information list 3701 as analysis candidate behavior information 3702. Note that if there are multiple pieces of analysis candidate behavior information 3602 having the earliest date in the normalized behavior period 3609, all of the analysis candidate behavior information 3602 that meet the above-mentioned selection criteria are added to the analysis candidate behavior information list 3701 as analysis candidate behavior information 3702. However, the analysis candidate behavior information 3602 to be added to the analysis candidate behavior information list 3701 may be narrowed down by any condition. As a result, zero or more pieces of analysis candidate behavior information 3602 are selected from the analysis candidate behavior information list 3601 as analysis candidate behavior information 3702 and added to the analysis candidate behavior information list 3701. When the control unit 11 of the sales activity analysis support system S completes the process in step S2709, the process proceeds to step S2710.

[0209] In step S2710, the control unit 11 of the sales activity analysis support system S causes the sales activity analysis unit 105 to perform the processes of steps S2704 to S2709 on all the behavior information 1602 that make up the behavior information list 1601, and then ends the loop process of steps S2703 to S2710.

[0210] In step S2711, the control unit 11 of the sales activity analysis support system S executes a process in which the sales behavior analysis unit 105 divides the analytical behavior information 3702 constituting the analytical behavior information list 3701 into multiple displayed behavior information groups 3808 by category. The division into multiple displayed behavior information groups 3808 may be performed based on any rule, or may be performed using a statistical model, a machine learning model, or a deep learning model, or a combination of these. In the third embodiment, first, analytical behavior information 3702 having the same subject position 3706 of the analytical behavior information 3702 is grouped into the same displayed behavior information group 3808. Then, using a deep learning model that inputs a character string and outputs a numerical string, the behavior content 3707 of the analytical behavior information 3702 constituting each displayed behavior information group 3808 is converted into a numerical string by the deep learning model. The numerical string is divided into multiple subcategories by unsupervised clustering into an arbitrary number of categories, thereby further dividing the single displayed behavior information group 3808. After the above-described division process, a list of analytical behavior information 3702 belonging to the subcategory is used as a display behavior information group 3808 in step S2712. A category name 3806 is assigned to each display behavior information group 3808 in consecutive order (e.g., Category 1). However, the category name 3806 may be determined based on the list of behavior details 3707 of the analytical behavior information 3702 that constitute the display behavior information group 3808. For example, the longest character string that appears most frequently in the list of behavior details 3707 may be used as the name of the category. However, the category name 3806 may be determined based on a different rule, or may be determined using a statistical model, a machine learning model, a deep learning model, or a combination of these. As a result, the analytical behavior information 3702 that constitutes the analytical behavior information list 3701 is divided into multiple display behavior information groups 3808 by category. After completing the process in step S2711, the control unit 11 of the sales activity analysis support system S proceeds to step S2712.

[0211] In step S2712, the control unit 11 of the sales activity analysis support system S executes processing to display an analysis screen 3801 by category using the sales behavior analysis unit 105. First, a list of the case IDs 3703 of the analytical behavior information 3702 constituting each display behavior information group 3808 is acquired. Based on the list of the case IDs 3703 of each display behavior information group 3808, the case management information list 2401 corresponding to the list of the case IDs 3703 is acquired, and the proportion of cases that have been accepted from the case management information list 2401 is calculated as the acceptance rate 3807 of each display behavior information group 3808. After the above processing, the specified information 3802 accepted in step S2701, the category name 3806 of each category, the acceptance rate 3807, and the display behavior information group 3808 are displayed in a layout similar to that of the analysis screen 3801. As a result, the analysis screen 3801 is displayed by category. When the process in step S2712 is completed, the control unit 11 of the sales activity analysis support system S ends the sales activity analysis process shown in the flowchart of FIG.

[0212] According to the third embodiment, it is possible to comprehensively and automatically record diverse and detailed behavioral details that could not be obtained with conventional techniques based on the contents of the input document 101, and to analyze sales activities based on diverse and detailed behaviors that are difficult to achieve with conventional techniques. For example, the analysis screen 3801 makes it possible to obtain insights into behavioral strategies for the success or failure of a sales case based on the diverse behaviors of stakeholders such as salespeople and customers.

[0213] The above-described embodiment of the present invention can be summarized as follows.

[0214] (1) The sales activity analysis support system S is a system for supporting the analysis of sales activities, and includes a behavior history database 106 that accumulates behavior information indicating behavior in sales activities for each sales case, and a case management database 107 that stores case management information for each sales case, including information on stakeholders in each sales case. Upon receiving one or more input documents 101 related to the sales case and identification information that uniquely identifies the sales case for which the input documents 101 were created, the system obtains case management information related to the sales case from the case management database 107 and stores the information in the behavior history database 106 at the time of receiving the input documents 101. The system includes an accumulated information summarizing unit 102 that acquires behavioral information related to sales cases stored in a database 106 and generates summary data based on the acquired behavioral information and case management information; a process estimation unit 103 that generates process information indicating the process up to the reception of the input document 101 based on the summary data and the input document 101; a behavior analysis unit 104 that generates zero or more pieces of new behavioral information based on the process information and the input document 101 and records the generated behavioral information in a behavior history database 106; and a sales behavior analysis unit 105 that performs an analysis of the sales activities based on the behavioral information stored in the behavior history database 106. This configuration enables the sales activity analysis support system S to comprehensively and automatically record diverse and detailed behavioral content that could not be acquired using conventional techniques based on the content of document data. This increases the variety of behavioral content that can be referenced in the analysis, enabling more advanced sales activity management, efficiency improvement, and strategy formulation. In other words, this configuration enables the sales activity analysis support system S to comprehensively and automatically acquire diverse and detailed behavioral content that could not be acquired using conventional techniques based on the content of document data, thereby enabling more advanced sales activity analysis. Furthermore, compared to a system in which a person reads and understands text data and directly inputs behavioral information into a database, the present invention realizes automatic acquisition of behavioral information, which also has the effect of improving work efficiency and reducing dependency on individuals.

[0215] (2) The behavior history database 106 stores behavior information of a business case up to an arbitrary past time point, and the history information is information indicating the history from a past time point to the time when the input document 101 is received.

[0216] (3) The behavioral information includes at least one character of text indicating the content of the behavior, information indicating the time when the behavior was performed, information indicating the subject of the behavior, and information indicating the subject's position in the sales case.

[0217] (4) The behavior analysis unit 104 generates behavior information using one or more of a statistical model, a machine learning model, and / or a deep learning model.

[0218] (5) The process estimation unit 103 generates process information using one or more of a statistical model, a machine learning model, and / or a deep learning model.

[0219] (6) Statistical models, machine learning models, and deep learning models are content-generating models.

[0220] (7) The device further includes a document type identification unit 108 that identifies document type information indicating the description format and / or type of the input document 101, and the process inference unit 103 generates process information indicating the process up to the reception of the input document 101 based on the input document 101, the document type information, and the summary data.

[0221] (8) Based on the input document 101 and the summary data, the process inference unit 103 obtains the process information of another similar case from the process history database 106, and generates process information indicating the process up to the time the input document 101 was received.

[0222] (9) The process inference unit 103 obtains the process information of another similar case from the process history database 106 based on the input document 101, the document type information, and the summary data, and generates process information indicating the process up to the reception of the input document 101.

[0223] (10) The behavioral information includes at least information indicating the reliability of the behavioral information.

[0224] (11) The sales behavior analysis unit 105 has a function of displaying a brief history of behavior in sales activities on the screen based on the behavior information accumulated in the behavior history database 106.

[0225] (12) The sales behavior analysis unit 105 has a function of displaying a screen for analyzing factors that lead to the success or failure of a sales case based on the behavior information accumulated in the behavior history database 106.

[0226] (13) Information about stakeholders includes information about the person in charge of the business and / or the customer representative.

[0227] The present invention is not limited to the above-described embodiments, and can be implemented using any components without departing from the spirit of the present invention.

[0228] Each of the above embodiments is merely an example, and the present invention is not limited to these contents as long as the characteristics of the invention are not impaired. Furthermore, although various embodiments have been described above, the present invention is not limited to these contents, and not all of these contents are necessarily essential to the solution of the present invention. Other embodiments conceivable within the scope of the technical idea of ​​the present invention are also included within the scope of the present invention.

[0229] In the above figures, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily show all the control lines and information lines that are necessary for implementation. For example, it can be assumed that in reality, almost all components are interconnected.

[0230] Furthermore, the layout of each functional unit of the sales activity analysis support system S described above is merely an example. The layout of each functional unit can be changed to an optimal layout in terms of the performance, processing efficiency, communication efficiency, etc. of the hardware and software provided in the sales activity analysis support system S.

[0231] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by having processor 1 interpret and execute programs that realize the respective functions. [Explanation of symbols]

[0232] 100...Sales activity analysis support system

Claims

1. A sales activity analysis support system that supports the analysis of sales activities, A behavioral history database that accumulates behavioral information indicating behavior in sales activities for each sales case; a project management database that stores project management information for each business project, including information about stakeholders in each business project; Equipped with an accumulated information summarizing unit that, upon receiving one or more input documents related to the business case and identification information that uniquely identifies the business case for which the input documents were created, acquires case management information related to the business case from the case management database, acquires behavior information related to the business case that is accumulated in the behavior history database at the time the input documents are received, and generates summary data based on the acquired behavior information and case management information; a process inference unit that generates process information indicating a process up to the reception of the input document based on the summary data and the input document; a behavior analysis unit that generates zero or more pieces of behavior information based on the context information and the input document and records the generated behavior information in the behavior history database; a sales behavior analysis unit that executes an analysis process of the sales activities based on the behavior information stored in the behavior history database; A sales activity analysis support system with the above.

2. 2. The sales activity analysis support system according to claim 1, The behavior history database stores behavior information of the sales case up to an arbitrary past point in time, The history information is information indicating a history from the past time point to the time point when the input document is received. Sales activity analysis support system.

3. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system in which the behavioral information includes at least one character of text indicating the content of the behavior, information indicating the time when the behavior was performed, information indicating the subject of the behavior, and information indicating the subject's position in the sales case.

4. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system in which the behavioral analysis unit generates the behavioral information using one or more of a statistical model, a machine learning model, and / or a deep learning model.

5. 5. The sales activity analysis support system according to claim 4, A sales activity analysis support system, wherein the statistical model, the machine learning model, and the deep learning model are content generation models.

6. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system in which the process estimation unit generates the process information using one or more of a statistical model, a machine learning model, and / or a deep learning model.

7. 7. The sales activity analysis support system according to claim 6, A sales activity analysis support system, wherein the statistical model, the machine learning model, and the deep learning model are content generation models.

8. 2. The sales activity analysis support system according to claim 1, a document type identification unit for identifying document type information indicating the description format and / or type of the input document; the process inference unit generates process information indicating a process up to the reception of the input document based on the input document, the document type information, and the summary data; Sales activity analysis support system.

9. 2. The sales activity analysis support system according to claim 1, the process inference unit acquires, based on the input document and the summary data, action information of another similar case from the action history database, and generates process information indicating the process up to the time the input document was received. Sales activity analysis support system.

10. 9. The sales activity analysis support system according to claim 8, the process inference unit acquires, based on the input document, the document type information, and the summary data, action information of another similar case from the action history database, and generates process information indicating the process up to the reception of the input document; Sales activity analysis support system.

11. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system, wherein the behavioral information includes at least information indicating the reliability of the behavioral information.

12. 2. The sales activity analysis support system according to claim 1, The sales activity analysis support system has a function in which the sales activity analysis unit displays a brief history of sales activity behavior on a screen based on the behavior information accumulated in the behavior history database.

13. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system in which the sales behavior analysis unit has a function of displaying a screen for analyzing factors that lead to the success or failure of a sales case based on the behavioral information stored in the behavior history database.

14. 2. The sales activity analysis support system according to claim 1, A sales activity analysis support system, wherein the information about the stakeholders includes information about the person in charge of the sales case and / or the person in charge of the customer.

15. A sales activity analysis support method for supporting the analysis of sales activities, comprising: A behavioral history database that accumulates behavioral information indicating behavior in sales activities for each sales case; a project management database that stores project management information for each business project, including information about stakeholders in each business project; A computer including at least a storage device storing the above and a processor, an accumulated information summarizing unit, upon receiving one or more input documents related to the sales case and identification information that uniquely identifies the sales case for which the input documents were created, acquires case management information related to the sales case from the case management database, and acquires behavior information related to the sales case that is accumulated in the behavior history database at the time the input documents are received, and generates summary data based on the acquired behavior information and case management information; a process inference unit generates process information indicating a process up to the reception of the input document based on the summary data and the input document; a behavior analysis unit generates zero or more pieces of behavior information based on the process information and the input document, and records the generated behavior information in the behavior history database; a sales behavior analysis unit that performs an analysis process of the sales activities based on the behavior information stored in the behavior history database; Sales activity analysis support method.

Images