Information processing systems, information processing methods, and programs

Abstract

To enable users to easily identify knowledge. [Solution] An information processing system comprising: an attribute value storage unit that stores attribute values ​​for each of a plurality of attributes associated with knowledge data; a job storage unit that stores information that identifies attributes related to a task in association with the task; a task designation unit that receives task designations from a user; an attribute identification unit that identifies attributes corresponding to the accepted task by referring to the job storage unit; an attribute value reception unit that receives attribute value designations from a user for at least one of the identified attributes; and a search unit that searches the attribute value storage unit for knowledge data corresponding to the accepted attribute value.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and a program.

Background Art

[0002] Patent Document 1 discloses a technique for outputting a business document with a high degree of suitability in the next phase of business by selecting a business case.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When trying to identify the knowledge desired by the user himself / herself as in the technique disclosed in Patent Document 1, it is difficult to identify when the number of knowledge is large.

[0005] The present invention has been made in view of such a background, and an object thereof is to provide a technique that enables a user to easily identify knowledge.

Means for Solving the Problems

[0006] The main invention of the present invention for solving the above problems is an information processing system comprising: an attribute value storage unit that stores attribute values ​​for each of a plurality of attributes associated with knowledge data; a job storage unit that stores information that identifies the attribute related to a business in association with the business; a business designation unit that receives business designations from a user; an attribute identification unit that identifies the attribute corresponding to the received business by referring to the job storage unit; an attribute value reception unit that receives an attribute value designation from the user for at least one of the identified attributes; and a search unit that searches the attribute value storage unit for knowledge data corresponding to the received attribute value.

[0007] Further issues and solutions disclosed in this application will be made clear in the section on embodiments of the invention and in the drawings. [Effects of the Invention]

[0008] According to the present invention, users can easily identify knowledge. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the overall configuration of the information processing system according to the first embodiment. [Figure 2] This figure shows an example of the hardware configuration of the management server 2 according to the first embodiment. [Figure 3] This figure shows an example of the software configuration of the management server 2 according to the first embodiment. [Figure 4] This diagram illustrates the operation of the management server 2 according to the first embodiment. [Figure 5] This figure shows an example of the overall configuration of the information processing system according to the second embodiment. [Figure 6] This figure shows an example of the hardware configuration of the management server 2 according to the second embodiment. [Figure 7] This figure shows an example of the software configuration of the management server 2 according to the second embodiment. [Figure 8]This diagram illustrates the operation of the management server 2 according to the second embodiment. [Modes for carrying out the invention]

[0010] <First Embodiment> The following describes an information processing system according to the first embodiment of the present invention. The information processing system of the first embodiment manages knowledge and aims to make it easy for users to search for what they need from a large amount of knowledge. Knowledge represents the knowledge acquired by employees within an organization. Knowledge is assigned attribute values ​​for multiple attributes, such as the size of the customer or the industry of the customer (size may be the number of people, such as 100, sales amount, such as 1 billion, and industry may be "finance," "information and communication," etc.). Users can search for knowledge by specifying attribute values ​​for the desired attributes. In the information processing system of the first embodiment, by specifying a task, the system can suggest to the user that they specify attribute values ​​for attributes closely related to that task, so that knowledge with attribute values ​​set for many attributes can be searched efficiently. The task may be a sales task or a knowledge search task. In terms of tasks, for example, in solution sales, this could include "I want to find past cases to identify customer problems," "I want to find similar cases to create a proposal," or "I want to find examples of failures in similar past cases." In account sales, it could include "I want to create materials to initiate business negotiations," "I want to create discussion materials that allow me to delve deeper into customer interests," or "I want to make appropriate cross-selling proposals tailored to their needs."

[0011] Figure 1 is a diagram showing an example of the overall configuration of an information processing system according to the first embodiment. The information processing system of the first embodiment is configured to include a management server 2. The management server 2 is connected to the user terminal 1 via a communication network. The communication network is, for example, the internet and is constructed using public telephone lines, mobile phone lines, wireless communication channels, Ethernet (registered trademark), etc.

[0012] User terminal 1 is a computer operated by the user. User terminal 1 can be, for example, a smartphone, a tablet computer, or a personal computer.

[0013] The management server 2 may be a general-purpose computer such as a workstation or personal computer, or it may be logically implemented through cloud computing.

[0014] <Management Server> Figure 2 shows an example of the hardware configuration of the management server 2 according to the first embodiment. Note that the illustrated configuration is just one example, and other configurations are also possible. The management server 2 includes a CPU 201, memory 202, storage device 203, communication interface 204, input device 205, and output device 206. The storage device 203 stores various data and programs, such as a hard disk drive, solid-state drive, or flash memory. The communication interface 204 is an interface for connecting to a communication network, such as an adapter for connecting to Ethernet®, a modem for connecting to a public telephone network, a wireless communication device for wireless communication, or a USB (Universal Serial Bus) connector or RS232C connector for serial communication. The input device 205 is for inputting data, such as a keyboard, mouse, touch panel, button, or microphone. The output device 206 is for outputting data, such as a display, printer, or speaker. Furthermore, each functional unit of the management server 2, as described later, is realized by the CPU 201 reading programs stored in the storage device 203 into memory 202 and executing them, and each storage unit of the management server 2 is realized as part of the storage area provided by memory 202 and storage device 203.

[0015] FIG. 3 is a diagram showing an example of the software configuration of the management server 2 according to the first embodiment. The management server 2 includes a knowledge storage unit 231, an attribute value storage unit 232, a job storage unit 233, a user information storage unit 234, a service designation unit 211, an attribute identification unit 212, an attribute value reception unit 213, a search unit 214, and an output unit 215.

[0016] <Storage unit> The knowledge storage unit 231 stores data representing knowledge (hereinafter referred to as knowledge data). The knowledge data can be, for example, files such as word processor files, presentation files, spreadsheet files, PDF files, or records managed in a database such as an RDBMS. The knowledge storage unit 231 may be implemented, for example, as a folder managed by a file system, or may be implemented by an RDBMS or an object database.

[0017] The attribute value storage unit 232 stores attribute values ​​associated with knowledge data. Knowledge data is associated with multiple attributes, and the attribute value storage unit 232 stores attribute values ​​for each of these multiple attributes. The attribute value storage unit 232 can store multiple knowledge data in association with each of the multiple attributes associated with each knowledge data and its attribute value. Attributes may include, for example, industry, company size, project type, internal department, and product / service field. Each attribute is associated with its attribute value; for example, the industry attribute may be associated with attribute values ​​such as "finance," "manufacturing," and "distribution," and the company size attribute may be associated with attribute values ​​such as "large company," "small and medium-sized enterprise," and "venture company." The attribute value storage unit 232 can store records (hereinafter referred to as attribute value information) that associate information identifying the knowledge data, information identifying the attribute, and the attribute value. The attribute value storage unit 232 can store, for example, information about a given piece of knowledge data, such as that the associated "customer size" is in units of "100 people" and that the associated "customer industry" is "information and communication." The attribute value storage unit 232 can store information that identifies an attribute (for example, text such as "customer size" or an attribute ID) and an attribute value (for example, "100 people," without including the unit) associated with information that identifies the piece of knowledge data (for example, a path, file name, URL, unique ID, etc.). The attribute value storage unit 232 can store attribute values ​​for multiple attributes for a single piece of knowledge data. Multiple attribute values ​​may be stored for a single piece of knowledge data and a single attribute. For example, the attribute value storage unit 232 may set attribute values ​​of "10 people" and "100 people" for the "customer size" attribute.

[0018] The job storage unit 233 stores information related to operations (hereinafter referred to as job information). The job information includes information for specifying an operation and information for specifying an attribute related to the operation, and the job storage unit 233 can store information for specifying an attribute related to the operation in association with the operation. The job storage unit 233 may store information for specifying a plurality of attributes in association with one operation. For example, a plurality of job information including information for specifying the same operation and information for specifying different attributes can be registered in the job storage unit 233.

[0019] The job storage unit 233 may store information for specifying an attribute related to the operation and industry in association with the operation and industry. In this case, the job information may include information for specifying an operation, information for specifying an industry, and information for specifying an attribute. Further, the job information may include a priority.

[0020] The user information storage unit 234 stores information related to a user (hereinafter referred to as user information). The user information includes information for specifying a user and information for specifying an industry in which the user is engaged, and the user information storage unit 234 can store the industry in which the user is engaged for each user.

[0021] <Functional unit> The operation specifying unit 211 receives a specification of an operation from the user. The operation specifying unit 211 can, for example, read out without duplication information for specifying an operation included in the job information from the job storage unit 233. The operation specifying unit 211 creates screen data for displaying a list of operations specified by the information for specifying the read operation, and transmits the created screen data to the user terminal 1 so that the user can select an operation from the list of operations on the user terminal 1. The operation specifying unit 211 may receive a specification of an operation as free-form text data. The operation specifying unit 211 can receive operation specification information transmitted from the user terminal 1. The operation specification information can be, for example, an operation ID, a name of an operation, a code for identifying an operation, a field of an operation, and the like.

[0022] The business designation unit 211 may accept the designation of a business type in addition to or instead of a business. The business designation unit 211 may also refer to the user information storage unit 234 to obtain the business type corresponding to the user from the information that identifies the business type included in the user information.

[0023] The attribute identification unit 212 identifies attributes related to a task. The attribute identification unit 212 identifies attributes related to a task received by the task designation unit 211. The attribute identification unit 212 can identify attributes corresponding to a task received by the task designation unit 211 by referring to the job storage unit 233. The attribute identification unit 212 can identify multiple attributes related to a task. The attribute identification unit 212 can set a priority for the identified tasks. In the first embodiment, the attribute identification unit 212 can use the priority included in the job information stored in the job storage unit 233 as the priority of the attributes.

[0024] Furthermore, the attribute identification unit 212 may be configured to identify the attributes corresponding to the accepted business and / or business when the business designation unit 211 receives a designation of a business type.

[0025] The attribute value receiving unit 213 receives attribute values ​​related to the knowledge desired by the user. The attribute value receiving unit 213 can receive attribute values ​​from the user terminal 1 for attributes specified by the business specification unit 211. Attribute values ​​can be received, for example, by having the user select from attribute values ​​stored in the attribute value storage unit 232, or by having the user input attribute values ​​into a text box, etc. The attribute value receiving unit 213 receives the user's specification of an attribute value for at least one of the attributes specified by the attribute identification unit 212. The attribute value receiving unit 213 can receive attribute values ​​sequentially in order of attribute priority. For example, it can receive attribute values ​​for the attribute with the highest priority, and then receive attribute values ​​for the attribute with the next highest priority. Alternatively, for example, screen data representing a form that displays input fields in order of priority may be sent to the user terminal 1. The attribute value receiving unit 213 may retrieve the attribute values ​​stored in the attribute value storage unit 232 for each specified attribute without duplication, output a list of attribute values ​​for each attribute, and allow the user to select an attribute value from that list. For example, when the attribute value receiving unit 213 receives an attribute value for the "customer size" attribute, it reads all the attribute values ​​set for the "customer size" attribute (e.g., "10 people", "50 people", "100 people", "500 people", "1000 people", etc.) from the attribute value storage unit 232. The attribute value receiving unit 213 then displays the read attribute values ​​in a list and allows the user to select an attribute value from that list.

[0026] The method by which the attribute value receiving unit 213 receives attribute values ​​is not limited to the above example. For example, the attribute value receiving unit 213 may accept direct input of attribute values ​​into a text box or the like, or it may display a list of attribute value candidates and allow the user to select from there. In addition, the attribute value receiving unit 213 may accept input using voice input or a pointing device.

[0027] The search unit 214 retrieves knowledge data corresponding to the received attribute value from the attribute value storage unit 232. The search unit 214 can use the attribute value received by the attribute value receiving unit 213 as a search key to search for knowledge data stored in the attribute value storage unit 232. For example, the search unit 214 can execute a query against the attribute value storage unit 232 using a database query language such as SQL (Structured Query Language) and obtain the query results. If the attribute value receiving unit 213 receives attribute values ​​one by one sequentially, the search unit 214 can add the received attribute values ​​as an AND condition and search for knowledge data that matches the condition (increment search).

[0028] The attribute value receiving unit 213 may accept input of search text data regardless of the attribute specified by the business specification unit 211. The search unit 214 may use the text data received by the attribute value receiving unit 213 to narrow down the search results. That is, the search unit 214 searches the attribute value storage unit 232 for knowledge data corresponding to the received attribute value, and can also search the search results for knowledge data in which the text data matches the content of the knowledge data, and / or the text data matches the attribute value of the attribute associated with the knowledge data. Text data matching a certain value may include the value containing text that matches or is similar to the text data, or the distance between the vector with the value embedded and the vector with the text data embedded is within a predetermined value. Furthermore, the search unit 214 can also combine the search results obtained by searching the attribute value storage unit 232 for knowledge data corresponding to the received attribute value, with the search results obtained by searching all knowledge data for text data that matches the content of the knowledge data, and / or for text data that matches the attribute value of an attribute associated with the knowledge data, to obtain the final search result.

[0029] The output unit 215 outputs the retrieved knowledge data. The output unit 215 can output a list of information that identifies the retrieved knowledge data. The output unit 215 can output information that identifies the retrieved knowledge data and a portion of the knowledge data. In this case, the output unit 215 can obtain the knowledge data retrieved by the search unit 214 from the knowledge data storage unit 231 and output a portion of the obtained knowledge data (which may be processed, such as by creating a thumbnail of a page image). The output unit 215 can, for example, create screen data for displaying a list of knowledge data or screen data for displaying knowledge data in detail and send it to the user terminal 1.

[0030] <Operation> Figure 4 is a diagram illustrating the operation of the management server 2.

[0031] Management server 2 receives a request for a task (S301), identifies the attribute corresponding to the specified task (S302), receives the attribute value for the attribute (S303), and searches for knowledge data corresponding to the attribute value (S304).

[0032] As described above, the information processing system of the first embodiment makes it possible to easily and efficiently identify knowledge related to business operations.

[0033] The first embodiment has been described above, but the above embodiment is for the purpose of facilitating understanding of the present invention and is not intended to limit the interpretation of the present invention. The present invention can be modified and improved without departing from its spirit, and equivalents thereof are also included in the present invention.

[0034] For example, the processing performed by each functional unit of the management server 2 described above may be executed by any of the functional units. Furthermore, different functional units may be added to perform some of the processing performed by each of the functional units described above. Also, the functional units of the management server 2 may be distributed across multiple computers.

[0035] Furthermore, the information stored in each memory unit of the management server 2 may be stored in any of the memory units. That is, the information stored in the multiple memory units mentioned above may be stored in a single memory unit, or a portion of the information stored in one memory unit may be stored in another memory unit.

[0036] <Example 1> In the first embodiment described above, the correspondence between business (and / or industry) and attributes is stored in advance. However, the attributes used for filtering may be stored in the job storage unit 233 by associating business (and / or industry) with attribute values ​​related to predetermined attributes (multiple attributes are possible). In this case, for the first attribute, the one corresponding to the business (and / or industry) is identified, but for the second and subsequent attributes, the next attribute corresponding to the business (and / or industry) and the attribute value for the attribute for which input was received can be identified. If an attribute corresponding to the business (and / or industry) and the attribute value for the attribute for which input was received is not stored, the attribute value for that attribute can be excluded from the search criteria and the search can be performed. This makes it possible, for example, to change the attribute to be queried next depending on whether the company size is large or small after the company size is initially selected.

[0037] <Modification 2> In the above embodiment, the correspondence between business (and / or industry) and attributes is stored in advance, but it may also be determined without prior preparation. The attribute identification unit 212 can, for example, infer relevant attributes from business and / or industry using a machine learning model.

[0038] For example, the management server 2 may include a learning model storage unit that stores a machine learning model created by machine learning using business (and / or industry) and related attributes as training data. The attribute identification unit 212 can identify one or more attributes related to a business (and / or industry) by providing the business (and / or industry) specified by the user to the machine learning model. In this case, the machine learning model may infer the degree to which each attribute is related to the business (and / or industry) and use that degree as a priority for businesses (and / or industries) that are equal to or greater than a predetermined value.

[0039] Alternatively, machine learning generators may be used to generate attributes related to the business (and / or industry). For example, a large-scale language model can be given a prompt that includes the business (and / or industry) and instructions to generate attributes related to that business (and / or industry) to generate attributes. In this case, the prompt may also include a list of attributes associated with knowledge data, and instructions to select from the list.

[0040] <Variation 3> The search unit 214 can search the knowledge data using all of the attribute values ​​received by the attribute value receiving unit 213 as search conditions. Alternatively, the search unit 214 may use only a portion of the received attribute values ​​as search conditions.

[0041] Furthermore, when the search unit 214 receives multiple attribute values, it may perform the search by combining each attribute value using an "OR" condition instead of an "AND" condition. For example, instead of the condition "Industry is finance" and "Deal type is investment," the search can be performed using the condition "Industry is finance" or "Deal type is investment." The user can specify whether to use an AND or OR condition.

[0042] <Modification 4> In the embodiments described above, the output unit 215 was described as outputting knowledge data retrieved by the search unit 214, but it is not limited to this. The output unit 215 may change the display order of the search results, taking into account the user's preferences and past selection history.

[0043] For example, the user information storage unit 234 may be configured to store the user's past search history and knowledge data selection history, and the output unit 215 may refer to the user information storage unit 234 to prioritize displaying knowledge data similar to the knowledge data previously selected by the user. The degree of similarity can be determined, for example, based on the degree of agreement of attribute values ​​assigned to the knowledge data or the degree of co-occurrence of keywords within the knowledge data.

[0044] Furthermore, the output unit 215 may change the display order of the search results in accordance with a display order change request sent from the user terminal 1. A display order change request is sent from the user terminal 1 when the user performs an operation to change the display order of the search results (for example, dragging and dropping items or clicking a button).

[0045] Furthermore, the output unit 215 may change the display order of search results according to the user's position and responsibilities within the company. For example, knowledge data related to strategic planning may be displayed preferentially for management users, and knowledge data related to business negotiations and contracts may be displayed preferentially for sales representative users. In this case, the user's position and responsibilities may be included in the user information stored in the user information storage unit 234.

[0046] <Modification 5> The output unit 215 can output the knowledge data retrieved by the search unit 214, for example, in descending order of search score. The output unit 215 may also change the display order of search results, taking into account the user's preferences and past selection history. For example, the user information storage unit 234 can store the user's past search history and the history of selected knowledge data, and the output unit 215 can refer to the user information storage unit 234 to display knowledge data containing attribute values ​​similar to previously selected knowledge data at the top of the search results.

[0047] Furthermore, the output unit 215 may refer to the user's attribute information (for example, department, job title, job duties, etc.) stored in the user information storage unit 234 and prioritize displaying knowledge data that matches the user's attributes.

[0048] Furthermore, the output unit 215 may use methods such as collaborative filtering to estimate the user's preferences and prioritize the display of knowledge data that matches the estimated preferences. When performing collaborative filtering, the selection history of other users stored in the user information storage unit 234 is referred to, and knowledge data selected by other users with similar preferences to the user in question is extracted.

[0049] <Variation 6> The attribute value storage unit 232 may also store additional information related to the attribute values. Here, additional information refers to information indicating the importance or reliability of the attribute value. For example, the attribute value storage unit 232 stores importance information for each attribute value, indicating the importance of the concept represented by that attribute value. The importance information can be represented, for example, by a numerical score (for example, a 5-point scale from 1 to 5).

[0050] Furthermore, the attribute value storage unit 232 may store confidence information for each attribute value, indicating the reliability of that attribute value. The confidence information can be assigned, for example, based on the reliability of the source of the attribute value. If the source is a reliable organization, the confidence level is set high, and if the source is unreliable, the confidence level is set low.

[0051] The search unit 214 can refer to additional information stored in the attribute value storage unit 232 and prioritize searching for knowledge data that includes attribute values ​​of high importance or high reliability. For example, the search unit 214 can calculate a search score for knowledge data based on the importance of the attribute values ​​and output the knowledge data in descending order of search score. The search unit 214 can also exclude knowledge data that includes attribute values ​​of low reliability from the search results or set a lower search score for such data.

[0052] <Example 7> The user information storage unit 234 can store information indicating the user's search history and preferences, in addition to attribute information such as the user's industry. For example, the user information storage unit 234 can store as search history the attribute values ​​the user has searched for in the past and the attribute values ​​attached to the knowledge data selected by the user. Furthermore, the user information storage unit 234 can generate and store information indicating the user's preferences by analyzing the trends in the attribute values ​​of the knowledge data selected by the user.

[0053] The attribute value receiving unit 213 can present attribute value candidates that are suitable for the user. For example, the attribute value receiving unit 213 may prioritize presenting attribute values ​​that the user has searched for in the past as candidates, based on the user's search history. Alternatively, the attribute value receiving unit 213 may prioritize presenting attribute values ​​that are presumed to be preferred by the user, based on the user's preferences. In this case, the attribute value receiving unit 213 may highlight attribute values ​​that have been searched for in the past or that match the user's preferences on the screen of the user terminal 1, or it may place these attribute values ​​at the top of the list. The attribute value receiving unit 213 may also exclude attribute values ​​other than those that have been searched for in the past or that match the user's preferences from the list of candidates.

[0054] <Differentiation Example 8> The user information storage unit 234 can store information that identifies the user and information that identifies the industry in which the user works, as well as information that indicates the user's search history and preferences. For example, the user information storage unit 234 stores keywords that the user has searched for in the past and knowledge data that the user has selected from the search results as part of the user's search history. The user information storage unit 234 also stores preference information that indicates what attribute values ​​the user prefers for each attribute. This preference information can be generated, for example, based on attribute values ​​specified by the user during a search or attribute values ​​that the user has selected from the search results.

[0055] The attribute value receiving unit 213 may refer to the user's search history and preference information stored in the user information storage unit 234 and present attribute value candidates suitable for the user. For example, the attribute value receiving unit 213 can present attribute values ​​previously specified by the user or attribute values ​​stored as preference information as attribute value candidates for the attribute the user is currently searching for. Alternatively, the attribute value receiving unit 213 may analyze the preference information of multiple users and present attribute values ​​common to many users as attribute value candidates.

[0056] <Modification 9> In the embodiment described above, the entire knowledge storage unit 231 is searched, but the search location may be narrowed down. In this case, the job storage unit 233 can store the business (and / or industry), information that identifies the search location (hereinafter referred to as "space"), and attributes in association with each other. The search unit 214 can also search for the location identified by the space (for example, if the knowledge storage unit 231 is a file system, this can be a specific directory. If the knowledge storage unit 231 is an object database or a relational database, each record can be made to include a space, and the space can be used as an AND condition for the search. This improves the efficiency of searching for business-related knowledge and suppresses the inclusion of noise.

[0057] <Disclosure Items> Furthermore, this disclosure also includes the following configurations. [Item 1] An attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, A job storage unit that stores information identifying the aforementioned attributes related to a task, in correspondence with the task, The task assignment department accepts task assignments from users, An attribute identification unit that identifies the attributes corresponding to the received business by referring to the job storage unit, An attribute value receiving unit that receives the specification of the attribute value from the user for at least one of the identified attributes, A search unit that retrieves the knowledge data corresponding to the received attribute value from the attribute value storage unit, An information processing system characterized by comprising the following features. [Item 2] The information processing system described in item 1, The attribute value receiving unit retrieves the attribute value stored in the attribute value storage unit without duplication for each of the identified attributes, and outputs a list of the attribute values ​​for each attribute. An information processing system characterized by the following. [Item 3] The information processing system described in item 1, Regardless of the attribute, the attribute value receiving unit receives text data input from the user. The search unit searches for knowledge data that corresponds to the received attribute value and in which the received text data matches at least one of the contents of the knowledge data or the attribute value of the attribute associated with the knowledge data. An information processing system characterized by the following. [Item 4] The information processing system described in item 1, The job storage unit stores information that identifies the attributes in association with the business and industry, The attribute value receiving unit receives the designation of the business and industry from the user, and identifies the attribute corresponding to the received business and industry. An information processing system characterized by the following. [Item 5] The information processing system described in item 1, The system includes a user information storage unit that stores the type of industry in which each user is engaged. The job storage unit stores information that identifies the attributes in association with the business and industry, The attribute value receiving unit refers to the user information storage unit to identify the industry corresponding to the user, and to identify the business specified by the user and the attribute corresponding to the industry corresponding to the user. An information processing system characterized by the following. [Item 6] The steps include storing attribute values ​​in an attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, The steps include: storing information in the job storage unit that identifies the attributes related to the business in accordance with the business; The steps include: referring to the job storage unit to identify the attributes corresponding to the task specified by the user; A step of receiving a specification of the attribute value from the user for at least one of the identified attributes, The steps include: searching the attribute value storage unit for the knowledge data corresponding to the received attribute value; An information processing method characterized by a computer executing the following. [Item 7] The steps include storing attribute values ​​in an attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, The steps include: storing information in the job storage unit that identifies the attributes related to the business in accordance with the business; The steps include: referring to the job storage unit to identify the attributes corresponding to the task specified by the user; A step of receiving a specification of the attribute value from the user for at least one of the identified attributes, The steps include: searching the attribute value storage unit for the knowledge data corresponding to the received attribute value; A program that causes a computer to execute something.

[0058] <Second Embodiment> The following describes an information processing system according to a second embodiment of the present invention. The information processing system according to the second embodiment identifies sales tasks (also called jobs; for example, product introductions, customer proposals, sales progress, etc.) performed by the user, and attempts to search for and provide knowledge using a search method that is optimal for those tasks.

[0059] Figure 5 shows an example of the overall configuration of an information processing system according to the second embodiment. The information processing system of the second embodiment includes a management server 2. The management server 2 is connected to the user terminal 1 via a communication network. The communication network is, for example, the internet and is constructed using public telephone lines, mobile phone lines, wireless communication channels, Ethernet (registered trademark), etc.

[0060] User terminal 1 is a computer operated by the user. User terminal 1 can be, for example, a smartphone, a tablet computer, or a personal computer.

[0061] The management server 2 may be a general-purpose computer such as a workstation or personal computer, or it may be logically implemented through cloud computing.

[0062] <Management Server> Figure 6 shows an example of the hardware configuration of the management server 2 according to the second embodiment. Note that the illustrated configuration is just one example, and other configurations are also possible. The management server 2 includes a CPU 201, memory 202, storage device 203, communication interface 204, input device 205, and output device 206. The storage device 203 stores various data and programs, such as a hard disk drive, solid-state drive, or flash memory. The communication interface 204 is an interface for connecting to a communication network, such as an adapter for connecting to Ethernet®, a modem for connecting to a public telephone network, a wireless communication device for wireless communication, or a USB (Universal Serial Bus) connector or RS232C connector for serial communication. The input device 205 is for inputting data, such as a keyboard, mouse, touch panel, button, or microphone. The output device 206 is for outputting data, such as a display, printer, or speaker. Furthermore, each functional unit of the management server 2, as described later, is realized by the CPU 201 reading programs stored in the storage device 203 into memory 202 and executing them, and each storage unit of the management server 2 is realized as part of the storage area provided by memory 202 and storage device 203.

[0063] Figure 7 shows an example of the software configuration of the management server 2. The management server 2 comprises a business identification unit 211, a search unit 212, a progress detection unit 213, a display unit 214, an input reception unit 215, a customer information acquisition unit 216, a knowledge classification unit 217, a recommendation generation unit 218, a knowledge storage unit 231, a suggested product storage unit 232, a search method storage unit 233, a customer information storage unit 234, a user information storage unit 235, a sales business storage unit 236, and a search history storage unit 237.

[0064] <Knowledge Memory Unit 231> The knowledge storage unit 231 is a database that stores multiple pieces of knowledge used in sales activities. The knowledge storage unit 231 can be implemented as, for example, a relational database, a document-oriented database, or a hybrid database combining these.

[0065] The knowledge storage unit 231 stores multiple types of knowledge used in various aspects of sales activities. Examples of knowledge include product introduction materials, competitor comparison materials, customer case studies, proposal templates, price lists, manuals, FAQs, and contract templates. The knowledge storage unit 231 also stores at least one of the customer's industry or customer challenges in association with the knowledge.

[0066] The knowledge storage unit 231 may include, for example, a knowledge table 231a, a knowledge classification table 231b, a business response table 231c, a customer attribute table 231d, and a space table 231e.

[0067] Knowledge table 231a is a table that stores basic information about each piece of knowledge. Knowledge table 231a includes fields such as information that identifies the piece of knowledge (e.g., knowledge ID), title, creation date, update date, content, and file path. The content field stores the text content directly if the knowledge is in text format, or the extracted text if the knowledge is in file format such as PDF. The file path field may store the path to the original file.

[0068] Knowledge classification table 231b is a table that stores classification information for each piece of knowledge. Knowledge classification table 231b includes fields such as Knowledge ID, Knowledge type (product introduction, customer proposal, sales progress, etc.), category, and subcategory.

[0069] The business correspondence table 231c is a table that stores the correspondence between knowledge and sales operations. The business correspondence table 231c includes fields such as Knowledge ID, information that identifies the operation (e.g., Business ID), and Relevance. The Business ID refers to an operation defined in the sales operation storage unit 236. Relevance is a numerical representation of the strength of the relevance of the knowledge to a specific sales operation.

[0070] The customer attribute table 231d is a table that stores the correspondence between knowledge and customer attributes (industry, issues, etc.). The customer attribute table 231d includes fields such as Knowledge ID, industry, issues, size, and job title.

[0071] Space table 231e is a table that stores information about the space to which a piece of knowledge belongs. Space table 231e includes fields such as space identification information (e.g., space ID), space name, description, creator, creation date, update date, and access permissions. It also includes a knowledge space correspondence table that stores the correspondence between knowledge IDs and space IDs. A single piece of knowledge can belong to multiple spaces.

[0072] Furthermore, the system includes a business space correspondence table that stores the degree of relevance between sales operations and spaces. The business space correspondence table includes fields such as business ID, space ID, and relevance. The relevance is a numerical representation of the strength of the relevance of a particular space to a specific sales operation. This relevance information is used by the search unit 212 to recommend spaces to search based on the sales operations identified by the business identification unit 211.

[0073] <Suggested product storage section 232> The proposed product storage unit 232 is a database that stores information on products that sales representatives should propose to customers. The proposed product storage unit 232 can be implemented, for example, as a relational database.

[0074] The suggested product storage unit 232 may include, for example, a product master table 232a, a suggested priority table 232b, and a customer-specific suggested product table 232c.

[0075] Product Master Table 232a is a table that stores basic information about all products offered by the company. Product Master Table 232a includes fields such as product identification information (e.g., product ID), product name, category, release date, price, features, and status (new product, standard product, old product, etc.). Technical information such as detailed specifications and feature lists for each product are also stored in association with it.

[0076] The Proposal Priority Table 232b is a table that stores information about the proposal priority of each product. The Proposal Priority Table 232b includes fields such as Product ID, Overall Priority, Industry-Specific Priority, and Recommendation Period. Overall Priority is a numerical value indicating the proposal priority of the product for all customers and is set based on business strategy and product lifecycle. Industry-Specific Priority is a numerical value indicating the proposal priority of the product for a specific industry. Recommendation Period indicates the period during which the proposal of the product is particularly recommended.

[0077] The customer-specific product suggestion table 232c is a table that stores information on products that should be suggested to a specific customer. The customer-specific product suggestion table 232c includes fields such as customer identification information (e.g., customer ID), product ID, suggestion reason, suggestion status, and planned suggestion date. The suggestion reason indicates the reason for suggesting the product to the customer (e.g., expiration of maintenance period for existing products, new introduction due to business expansion, etc.). The suggestion status indicates the progress of the suggestion (not yet suggested, suggested, under consideration, etc.). The planned suggestion date indicates the date on which the suggestion is planned.

[0078] <Search method memory unit 233> The search method storage unit 233 is a database that stores the optimal search method according to the type of sales operation. In this embodiment, the search method storage unit 233 may be implemented as, for example, a relational database and may operate in cooperation with the sales operation storage unit 236.

[0079] The search method storage unit 233 may include a search method master table 233a, a business-specific search method table 233b, a search method parameter table 233c, and a search method parameter table 233c.

[0080] The search method master table 233a is a table that stores basic information about the search methods available in the system. The search method master table 233a includes fields such as information identifying the search method (e.g., search method ID), search method name, description, and execution priority. Examples of search methods include free word search, category search, chat search, recommendation display, page-level search, and space-specified search. Each search method may be accompanied by a description of its characteristics and uses.

[0081] The business-specific search method table 233b is a table that stores the correspondence between sales operations and search methods. The business-specific search method table 233b can include fields such as business ID, search method ID, application priority, and valid flag. The business ID refers to a business defined in the sales operation storage unit 236. The application priority is a numerical value indicating the priority of application of the search method to a particular sales operation, and is used to determine which search method to use preferentially when multiple search methods are applicable. The valid flag is a flag that indicates whether the search method is currently available.

[0082] The search method parameter table 233c is a table that stores parameter information used when executing each search method. The search method parameter table 233c can include fields such as search method ID, parameter name, default value, value range, and description. It also includes a business-specific default value table that stores the default values ​​of search parameters for each sales business. The business-specific default value table includes fields such as business ID, search method ID, parameter name, and default value. This allows for flexible setting of different execution parameters for each search method, as well as providing optimal search conditions as default values ​​for each sales business.

[0083] The search method parameter table 233c is a table that stores parameter information used when each search method is executed. The search method parameter table 233c may include fields such as search method ID, parameter name, default value, value range, and description.

[0084] Furthermore, a search criteria table 233d is included, which stores search criteria information for each sales task. The search criteria table includes fields such as task ID, search criteria name, search criteria value, and priority. This allows not only to select a search method, but also to apply different search criteria for each sales task. For example, in the "product introduction" task, product name and product category can be set as priority search criteria, while in the "customer proposal" task, customer industry and challenges can be set as priority search criteria. Users can also modify these default search criteria when executing a search.

[0085] <Customer information storage unit 234> The customer information storage unit 234 is a database that stores information about customers who are the target of sales activities. The customer information storage unit 234 can be implemented as, for example, a relational database and can operate in cooperation with the proposed product storage unit 232 and the knowledge storage unit 231.

[0086] The customer information storage unit 234 may include a customer basic information table 234a, a customer attribute table 234b, a customer issue table 234c, and a customer contact history table 234d.

[0087] Customer Basic Information Table 234a is a table that stores basic customer information. Customer Basic Information Table 234a includes fields such as Customer ID, Customer Name, Location, Year of Establishment, Number of Employees, Annual Sales, and Transaction Start Date. This basic information can be used to identify customers and understand their basic company profile.

[0088] The customer attribute table 234b is a table that stores customer attribute information. The customer attribute table 234b includes fields such as customer ID, industry, industry details, company size classification, growth phase, and transaction status. The industry classification records both broad categories (manufacturing, finance, services, etc.) and detailed classifications. The company size classification indicates categories such as large companies, medium-sized companies, and small and medium-sized enterprises. The growth phase indicates categories such as startups, growth phases, maturity phases, and restructuring phases. The transaction status indicates categories such as prospects, new customers, existing customers, and dormant customers.

[0089] The Customer Issues table 234c is a table that stores information about issues that customers are facing. The Customer Issues table 234c includes fields such as Customer ID, information identifying the customer's issue (e.g., Issue ID), Issue Category, Issue Details, Priority, Status, Registration Date, and Update Date. The Issue Category indicates broad classifications such as business efficiency improvements, cost reduction, sales expansion, and risk management. Issue Details is a field that describes the specific content of the issue. Priority is a numerical value indicating the importance or urgency of the issue. Status indicates the progress, such as not yet addressed, in progress, or resolved.

[0090] The Customer Contact History Table 234d is a table that stores the contact history with customers. The Customer Contact History Table 234d includes fields such as information that identifies the contact history record (e.g., History ID), Customer ID, Contact Date and Time, Contact Method, Person in Charge ID, Content, Result, and Next Action. The Contact Method indicates the means by which the contact was made, such as visit, phone call, email, or web conference. The Content field records the purpose of the contact and what was discussed. The Result field records the outcome or response of the contact. The Next Action field records the next action or plan to be taken.

[0091] <User information storage unit 235> The user information storage unit 235 is a database that stores information about users, such as sales representatives. The user information storage unit 235 can be implemented, for example, as a relational database and can operate in cooperation with the customer information storage unit 234 and the search history storage unit 237.

[0092] The user information storage unit 235 may include a user basic information table 235a, a user attribute table 235b, a customer assigned table 235c, and a user activity history table 235d.

[0093] User Basic Information Table 235a is a table that stores basic user information. User Basic Information Table 235a includes fields such as User ID, Name, Department, Position, Contact Information, Date of Employment, and Date of System Use Start. This basic information is used to identify users and manage system usage rights and notification recipients.

[0094] User Attribute Table 235b is a table that stores user attribute information. User Attribute Table 235b includes fields such as User ID, Area of ​​Expertise, Preferred Industry, Preferred Products, Years of Sales Experience, and Skill Level. Area of ​​Expertise indicates the area in which the user has particular expertise (such as product category or business area). Preferred Industry indicates the customer industry in which the user is most proficient. Preferred Products indicates the company's products in which the user is most proficient. Years of Sales Experience indicates the number of years of experience the user has as a salesperson. Skill Level is a numerical representation of the user's proficiency in sales skills.

[0095] The assigned customer table 235c is a table that stores information about the customers that a user is responsible for. The assigned customer table 235c includes fields such as User ID, Customer ID, Assignment Start Date, Assignment Category, and Assignment Priority. The Assignment Category indicates categories such as primary, secondary, and support. The Assignment Priority is a numerical value that indicates the priority when a user is responsible for multiple customers.

[0096] The User Activity History Table 235d is a table that stores a user's system usage activity history. The User Activity History Table 235d includes fields such as information identifying the activity (e.g., Activity ID), User ID, Activity Date and Time, Activity Type, Information identifying the knowledge or customer targeted by the activity (e.g., Target ID), Content, and Results. The Activity Type indicates the type of system operation, such as search, view, edit, download, or share. The Target ID is either a Knowledge ID indicating the knowledge targeted by the activity or a Customer ID indicating the customer. The Content field records the detailed content of the activity. The Results field records the results and effects of the activity.

[0097] The information stored in the user information storage unit 235 is used by the business identification unit 211, the search unit 212, and the recommendation generation unit 218 for appropriate knowledge retrieval and recommendation generation. For example, search results are prioritized and filtered based on the user's area of ​​expertise and preferred industry. For instance, a user specializing in manufacturing will be given priority in displaying knowledge related to manufacturing. Furthermore, the level of detail and difficulty of the knowledge provided may be adjusted according to the user's skill level. Less experienced users will be given priority in basic explanatory materials and sales procedures, while experienced users will be given priority in advanced technical materials and detailed analyses of success stories. In addition, the user's activity history can be used as important data for analyzing the usage and effectiveness of knowledge. In particular, if a particular piece of knowledge is used by many users or leads to a high conversion rate, the probability of it being recommended to other users in similar situations increases.

[0098] <Sales Operations Memory Unit 236> The sales operations memory unit 236 is a database that stores the definitions and structures of various tasks performed in sales activities. The sales operations memory unit 236 can be implemented, for example, as a relational database and can operate in conjunction with the search method memory unit 233 and the knowledge memory unit 231.

[0099] The sales operations storage unit 236 may include a sales operations master table 236a, a sales process table 236b, a business characteristics table 236c, and an inter-business relationship table 236d.

[0100] Sales Operations Master Table 236a is a table that stores basic information about sales operations. Sales Operations Master Table 236a includes fields such as Operation ID, Operation Name, Operation Category, Description, Purpose, and Target Audience. The Operation Category indicates major classifications such as sales preparation, customer contact, needs assessment, proposal, closing, and follow-up. The Description field describes the overview and implementation method of the operation. The Purpose field describes the achievement goals of the operation. The Target Audience indicates the customer-side personnel (decision-makers, practical staff, etc.) who are the main targets of the operation.

[0101] The sales process table 236b is a table that stores the position of each task within the overall sales process. The sales process table 236b includes fields such as information identifying the sales process (hereinafter referred to as process ID), task ID, sequence, prerequisites, completion conditions, and standard duration. Sequence is a numerical value indicating the order in which tasks are performed within the sales process. Prerequisites is a field that describes the conditions that must be met in order to start the task. Completion conditions is a field that describes the conditions for determining that the task has been completed. Standard duration indicates the standard time required to complete the task.

[0102] The Business Characteristics Table 236c is a table that stores characteristic information for each sales business. The Business Characteristics Table 236c includes fields such as Business ID, Required Skills, Importance, Difficulty, Success Factors, and Failure Risk. Required Skills indicate the skill set necessary to effectively perform the business. Importance quantifies the importance of the business within the overall sales process. Difficulty quantifies the difficulty of performing the business. Success Factors is a field that describes the important elements for the success of the business. Failure Risk is a field that describes the elements that could cause the business to fail.

[0103] The inter-business relationship table 236d is a table that stores the relationships between different sales operations. The inter-business relationship table 236d includes fields such as Business ID1, Business ID2, Relationship Type, and Relationship Strength. The Relationship Type indicates the type of relationship, such as contextual, inclusion, or complementary. Relationship Strength is a numerical representation of the strength of the relationship between the operations.

[0104] <Search history storage unit 237> The search history storage unit 237 is a database that stores the history of user search activities. The search history storage unit 237 can be implemented, for example, as a relational database and can operate in cooperation with the user information storage unit 235 and the knowledge storage unit 231.

[0105] The search history storage unit 237 may include a basic search history table 237a, a search criteria table 237b, a search results table 237c, and a search effectiveness measurement table 237d.

[0106] The search history basic table 237a is a table that stores basic information about search activities. The search history basic table 237a may include fields such as information that identifies a search (e.g., search ID), information that identifies the user who performed the search (e.g., user ID), search date and time, business ID, customer ID, search method ID, and session ID. The session ID is an identifier used to group a series of consecutive search activities. The search method ID refers to a search method defined in the search method storage unit 233.

[0107] The search criteria table 237b is a table that stores the detailed search criteria for each search. The search criteria table 237b can include fields such as search ID, condition item name, condition value, condition operator, and priority. The condition item name indicates the field to be searched (title, content, category, etc.). The condition value indicates the value specified as the search criterion. The condition operator indicates how the search criterion is applied (exact match, partial match, range specification, etc.). The priority is a numerical value that indicates the priority when there are multiple conditions.

[0108] The search results table 237c is a table that stores the results obtained by a search. The search results table 237c can include fields such as search ID, result rank, knowledge ID, display status, view status, and usage status. Result rank is a numerical value indicating the display rank within the search results. Display status is a flag indicating whether the search result was actually displayed to the user. View status is a flag indicating whether the user opened and viewed the search result. Usage status is a flag indicating whether the user actually used the search result in their work.

[0109] The search effectiveness measurement table 237d is a table that stores information for measuring the effectiveness of search activities. The search effectiveness measurement table 237d can include fields such as search ID, search time, number of search results, number of viewed knowledge items, number of used knowledge items, satisfaction rating, and business results. Search time indicates the time required to execute the search. Number of search results indicates the total number of results obtained by the search. Number of viewed knowledge items and number of used knowledge items indicate the number of knowledge items from the search results that were actually viewed or used. Satisfaction rating indicates the user's satisfaction rating (e.g., 5-point scale) of the search results. Business results is a field that records how the knowledge obtained by the search led to results in actual sales operations.

[0110] The information stored in the search history storage unit 237 can be used, for example, for the recommendation generation unit 218 to generate appropriate knowledge recommendations. For example, it can be used as reference information to prioritize recommending knowledge that other users have given high ratings to for similar sales operations or customer attributes. The information stored in the search history storage unit 237 can also be used to improve the search accuracy of the search unit 212. By analyzing the relationship between past search patterns and results, it can be used as learning data to display more appropriate knowledge at the top for similar search conditions. Furthermore, the information stored in the search history storage unit 237 can be used for the improvement and optimization of the entire system. By analyzing which search methods are highly effective for specific sales operations and what kind of knowledge actually leads to sales results, it can be used to optimize the search method storage unit 233 and the knowledge storage unit 231.

[0111] <Business Specification Department 211> The task identification unit 211 identifies the sales tasks performed by the user. The task identification unit 211 can identify sales tasks by collecting and analyzing data from multiple sources. For example, the task identification unit 211 can monitor user operations (button clicks, menu selections, keyword inputs, etc.) received via the input reception unit 215 and identify the current business context from the operation patterns. For example, if a user is browsing the "product information" category or searching with keywords such as "tell me about the product features," it can be determined that there is a high probability that the user is performing a "product introduction" task.

[0112] Furthermore, the business identification unit 211 can comprehensively analyze information such as customer information currently being handled, obtained via the customer information acquisition unit 216, the progress status of sales activities obtained from the progress detection unit 213, and the type of knowledge used immediately prior to the transaction. For example, the business identification unit 211 can determine that if it is in the initial visit stage for a new customer in the manufacturing industry, it is highly likely that it is performing a "product introduction" task. It can also determine that if the next system update date for an existing customer is approaching, it is highly likely that it is performing a "customer proposal" task, and if it is handling customer support after an order has been placed, it is highly likely that it is performing a "sales progress" task.

[0113] The task identification unit 211 integrates this information to estimate which task is most likely to be performed by the user from among the tasks defined in the sales task memory unit 236. Rule-based reasoning and machine learning can be used for this estimation. Rule-based reasoning determines the task based on the correspondence between clear system operations and tasks. For example, if the user selects the "Customer Proposal" menu, it can be determined to be the "Customer Proposal" task. Machine learning estimation can probabilistically estimate the task based on a model that has learned the relationship between past operation patterns and tasks.

[0114] The task identification unit 211 can also manage the progress of identified sales tasks. Based on the task execution sequence, prerequisites, and completion conditions defined in the sales task memory unit 236, it tracks the progress of the current task. For example, when a "customer proposal" task is started, it checks whether the prerequisite, "clarification of customer issues," has been completed, and prompts the user to complete the prerequisite task if necessary. In the case of a "sales progress" task, it manages the progress of each step, such as contract creation, approval application, and delivery plan formulation.

[0115] The task identification unit 211 can identify tasks, for example, when a user logs into the system, when a user explicitly selects a sales task (for example, when "Customer Proposal" is selected from the menu), when the user's actions or the customer being handled change, when a certain amount of time has elapsed (for example, 15 minutes), or when the progress detection unit 213 notifies the user of a change in sales progress.

[0116] The task identification unit 211 provides the search unit 212 with information on the identified sales tasks, enabling the selection of the most suitable search method for the characteristics of the task. For example, if the task is identified as "product introduction," the search unit 212 can select a search method that combines free word search and recommendation display, or it can select category search for the "customer proposal" task, or chat search for the "sales progress" task.

[0117] In addition to the automatic identification described above, the task identification unit 211 also allows users to explicitly specify tasks. For example, the task identification unit 211 can display task buttons such as "Product Introduction," "Customer Proposal," and "Sales Progress" at the top of the screen, allowing users to explicitly specify the task they are currently performing by clicking these buttons. This explicit specification can be set to take precedence over the automatic estimation by the task identification unit 211.

[0118] The task identification unit 211 may learn user usage patterns and continuously improve the accuracy of task estimation. The task identification unit 211 can analyze the relationship between each user's past operation patterns and selected tasks and perform task estimation adapted to the behavioral characteristics of individual users.

[0119] <Search section 212> The search unit 212 searches for knowledge. The search unit 212 searches for knowledge from the knowledge storage unit 231 using a search method corresponding to the sales business identified by the business identification unit 211. The search unit 212 executes the search process by referring to databases such as the knowledge storage unit 231, the proposed product storage unit 232, the search method storage unit 233, the customer information storage unit 234, and the user information storage unit 235.

[0120] The search unit 212 receives information on sales operations identified from the business identification unit 211. The search unit 212 refers to the search method storage unit 233 to identify the search method corresponding to that sales operation. The search method storage unit 233 defines the correspondence between each sales operation and a search method (free word search, category search, chat search, recommendation display, space-specific search, etc.), and the search unit 212 can select a search method based on this definition. In addition, the search unit 212 can refer to the business space correspondence table to identify spaces that are highly relevant to the sales operation and recommend them as search targets.

[0121] The search unit 212 can optimize not only the search method but also the search conditions based on the sales operations identified by the business identification unit 211. By referring to the business-specific search conditions table 233d, it sets search conditions (search keywords, filter conditions, sort order, etc.) suitable for the sales operation as default values. This saves the user the trouble of specifying search conditions from scratch and allows them to quickly obtain search results that are appropriate to the business context.

[0122] The search unit 212 can improve search accuracy by recommending the most suitable space for each sales task as the search target. By narrowing the search scope to spaces highly relevant to the business context, search noise is reduced, and search results that better match the user's intent can be provided. For example, in a search for "delivery date" in a "sales progress" task, narrowing the search to the "contract document space" allows for the accurate extraction of knowledge related to actual contractual delivery conditions, rather than just mentions of delivery dates in product descriptions.

[0123] Depending on the identified business operations, the search unit 212 can perform different search processes as follows:

[0124] (1) In the case of product introduction work The search unit 212 refers to the suggested product storage unit 232 to identify products that should be suggested to the currently serving customer. Based on customer information (customer ID, industry, size, etc.) obtained from the customer information acquisition unit 216, the search unit 212 can refer to the suggestion priority and customer-specific suggested product information stored in the suggested product storage unit 232 to identify the most suitable suggested product for that customer. The search unit 212 searches the knowledge storage unit 231 for knowledge related to the identified suggested product. At this time, a search method combining free word search and recommendation display can be mainly used. In free word search, related knowledge can be broadly searched using product names or keywords as search conditions. On the other hand, in recommendation display, in cooperation with the recommendation generation unit 218, knowledge that is particularly relevant to the customer or product is displayed preferentially.

[0125] For example, in product introduction operations, the search unit 212 can set the following search conditions as default values: • Search fields: Product name, product overview, features • Document types: Catalogs, datasheets, comparison tables Sort by: Latest Update Date • Filters: Current products only

[0126] (2) In the case of customer proposal work The search unit 212 receives the customer's industry and challenges from the user via the input reception unit 215. It also utilizes customer information obtained from the customer information acquisition unit 216. Based on this information, the search unit 212 can search the knowledge storage unit 231 for knowledge corresponding to the specified industry and challenges. In this case, category search can be mainly used. With category search, the user can efficiently search for relevant knowledge by selecting categories such as industry (manufacturing, finance, service, etc.) and challenges (business efficiency improvement, cost reduction, sales expansion, etc.). Each piece of knowledge in the knowledge storage unit 231 is assigned industry and challenge tags, and the search unit 212 can perform searches using these tags.

[0127] In customer proposal work, the following search criteria may be set as default values: • Search fields: Problem, solution, effect Document types: Proposal templates, case studies, solution overviews Sort by: Relevance • Filter: Examples that match the customer's industry.

[0128] (3) In the case of sales management work The search unit 212 can search for relevant knowledge based on user questions and instructions received via the input reception unit 215. In this process, chat search is primarily used. Chat search analyzes the user's natural language questions, understands their intent, and then searches for appropriate knowledge. For example, in response to the question, "Can you provide a contract template?", it searches for knowledge related to contract templates. Similarly, in response to the question, "Can you explain the approval process for this customer?", it searches for knowledge related to the approval process tailored to the customer's industry and size.

[0129] In the "Sales Progress" task, the following search criteria may be set as default values: • Search fields: Procedures, contract terms, internal approval processes Document types: contract templates, procedure manuals, checklists • Sort order: Process step order • Filter: Latest version only

[0130] These default search criteria are displayed before the user performs a search and can be modified as needed. Furthermore, if a user frequently uses certain criteria, the system can learn these trends and suggest personalized, optimal search criteria.

[0131] In sales operations, chat searches are provided with an AI-powered interactive search interface. This interface pre-displays frequently asked questions and instructions by default, allowing users to easily start searches by simply clicking on them. For example, standard questions such as "What are the contract procedures for this customer?", "Show me a delivery schedule template," and "What are the steps for issuing an invoice?" are displayed as buttons, and users can select these instead of typing them themselves. This allows users to quickly access the necessary information while saving the effort of typing questions in natural language.

[0132] The chat search utilizes artificial intelligence (AI) technology to understand the user's question intent and search for and present the most relevant knowledge. The AI ​​chat interface displays context-sensitive common questions and instructions as options before the user even types a question. These default questions are generated based on the following sources: • Typical question patterns related to identified sales activities • Questions frequently asked by other users in a similar business context • Specific questions related to the customer or product currently being addressed. • The user's own past question patterns

[0133] Users can select from these question buttons or enter their own custom questions. The AI ​​analyzes the selected or entered question, searches for appropriate knowledge, and presents the results in natural language. For particularly complex questions, it can also integrate relevant knowledge to generate an answer. Additional suggested questions related to the answer are also presented, allowing the user to delve deeper into the necessary information while continuing the conversation.

[0134] In addition to the basic search methods described above, the search unit 212 also provides the following advanced search functions.

[0135] (1) Page-level search The search unit 212 has the function of searching the knowledge stored in the knowledge storage unit 231 on a page-by-page basis, rather than on a document-by-document basis. In particular, for documents containing a large number of pages, it is possible to efficiently search and display only the specific pages that contain the information the user needs.

[0136] (2) Combined search The search unit 212 has the function of performing complex searches that combine multiple search methods. For example, it is possible to perform a step-by-step search by first narrowing down the industry using a category search and then applying a free-word search to the results.

[0137] (3) Context-aware search The search unit 212 has the function of performing searches that take into account the user's past search history and the current business context. It refers to past search information stored in the search history storage unit 237 and displays search results that are relevant to the current context as a priority.

[0138] (4) Personalized search The search unit 212 has a function to personalize search results based on the user's attributes (specialty field, preferred industry, years of sales experience, etc.) stored in the user information storage unit 235. For example, it prioritizes displaying manufacturing-related knowledge to users who specialize in manufacturing, and financial-related knowledge to users who specialize in finance.

[0139] (5) Search by specifying spaces The search unit 212 has the function of performing a search by specifying a particular space. Users can explicitly select a space to limit the search range. In addition, the search unit 212 can automatically recommend highly relevant spaces based on the sales tasks identified by the task identification unit 211. For example, if the "product introduction" task is identified, the "product catalog space" will be preferentially recommended; if the "customer proposal" task is identified, the "proposal example space" will be preferred; and if the "sales progress" task is identified, the "contract document space" will be preferred. The recommended spaces are set as the default values ​​for the search conditions, but users can change the search target spaces as needed.

[0140] A major advantage of space-based searches is that they improve search accuracy by narrowing the search scope to highly relevant knowledge. For example, searching all knowledge with the keyword "contract" will include mentions of "contract" in various contexts, such as references to contracts in product catalogs or contract terms in proposals. However, by narrowing the search to the "contract document space," only knowledge related to actual contract templates and contract procedures will be extracted, allowing users to access the desired knowledge more quickly and reliably.

[0141] Furthermore, space-based searches are effective even when the same keyword has different meanings in different business contexts. For example, the keyword "proposal" often refers to the product's proposal function in a "product introduction" task, but often refers to the proposal document or the proposal activity itself in a "customer proposal" task. By specifying the appropriate space, search results that match the meaning in the user's current business context can be prioritized. In this way, space-based searches do not merely limit the search scope, but improve search accuracy according to the user's business context.

[0142] The search unit 212 works in conjunction with the knowledge classification unit 217 to classify and display search results according to the type and purpose of the knowledge. For example, by classifying and displaying search results into categories such as "product description materials," "case studies," and "technical specifications," users can efficiently find the information they need.

[0143] Furthermore, the search unit 212 works in conjunction with the recommendation generation unit 218 to provide not only search results but also related recommendation knowledge. For example, when searching for product information materials, it can also recommend related customer case studies and comparison tables to support the user's information gathering.

[0144] The search results from the search unit 212 are presented to the user via the display unit 214. The display format of the search results is also optimized for each sales task; for example, thumbnail display is used for product introduction tasks, category-based list display for customer proposal tasks, and chat-style display for sales progress tasks.

[0145] <Progress detection unit 213> The progress detection unit 213 is a processing unit responsible for detecting the progress of the user's sales activities. The progress detection unit 213 can comprehensively detect the progress of sales activities using, for example, multiple information sources and pattern recognition algorithms. For example, it can collect and analyze information related to sales progress from the following information sources:

[0146] (1) User operation history The progress detection unit 213 analyzes the user's operation history recorded via the input reception unit 215. For example, it can infer the current sales phase from the operation history, such as which knowledge base articles were viewed, what information was searched for, and what documents were sent to customers. If there is a lot of viewing of product materials, it can be determined that the user is likely in the "product introduction" phase; if proposal templates are used, it is likely in the "customer proposal" phase; and if contract-related document operations are performed, it is likely in the "sales progress" phase.

[0147] (2) Contact history with customers The progress detection unit 213 can refer to the customer contact history table 234d stored in the customer information storage unit 234 and analyze the content and results of recent communications with the customer. For example, a contact record of "first visit" may indicate the initial phase, a record of "needs hearing completed" may indicate the intermediate phase, and records of "proposal submitted" or "quote sent" may indicate the later phase.

[0148] (3) Knowledge utilization status: The progress detection unit 213 analyzes the usage of knowledge recorded in the search history storage unit 237. By tracking in chronological order what kind of knowledge is used for a particular customer or deal, it is possible to infer the progress of the sales process. For example, the use of knowledge in the order of product catalogs, case studies, proposal templates, price lists, and contract templates indicates the natural progression of the sales process.

[0149] (4) Explicit milestones of sales activities The progress detection unit 213 refers to milestone information of sales activities explicitly recorded by the user. For example, it can detect status changes such as "needs confirmed," "proposal submitted," and "contract concluded," as well as updates to the deal status. This explicit progress information indicates the current progress with higher reliability than other indirect sources.

[0150] (5) Time elapsed The progress detection unit 213 considers the elapsed time since the start of sales activities and the elapsed time since the last status update. By comparing this with the standard phase period defined in the sales operations memory unit 236, it can determine whether the current progress is ahead of or behind schedule.

[0151] The progress information detected by the progress detection unit 213 can be used as follows.

[0152] (1) Support for specific tasks The detected progress information is provided to the business identification unit 211, which can serve as an important clue for identifying the user's current sales activities. For example, the business identification unit 211 can estimate that the activity is a "product introduction" activity in the early stages of the sales process, a "customer proposal" activity in the middle stages, and a "sales progress" activity in the later stages.

[0153] (2) Selection of the optimal search method Based on the progress, the search unit 212 can assist in selecting the optimal search method. For example, during the proposal creation stage, searching for proposal templates and similar cases may be prioritized, while during the contract negotiation stage, searching for knowledge related to contract terms and procedures may be prioritized.

[0154] (3) Visualization of progress The detected progress information can be visualized to the user through the display unit 214. For example, the current sales phase and progress rate can be displayed graphically, allowing the user to grasp the progress of their sales activities at a glance.

[0155] (4) Recommended actions Based on the progress, the next sales activity to be performed is recommended. For example, once the needs assessment is complete, creating and submitting a proposal is recommended as the next step. This allows the user to efficiently manage the sales process.

[0156] The progress detection unit 213 can also utilize machine learning technology to improve detection accuracy. Progress detection becomes possible by learning the relationship between past sales activity patterns and actual progress status. In particular, it can learn the sales style of individual users and the characteristics of sales processes for each customer industry to achieve personalized progress detection.

[0157] Furthermore, the progress detection unit 213 also has a function to detect abnormalities or stagnation in sales activities. If the progress pattern deviates significantly from the standard pattern, or if the progress has not been updated for a long period of time, it can be detected as a potential problem and the user can be notified.

[0158] <Display section 214> The display unit 214 is a processing unit responsible for presenting knowledge and related information retrieved by the search unit 212 to the user. In the second embodiment, the display unit 214 is configured as a software module implemented on the server and client terminals of the information processing system.

[0159] The main function of the display unit 214 is to provide a user interface optimized according to the type of sales work identified by the business identification unit 211. In particular, the display format of search results and the priority of information change depending on the type of sales work, so that users can obtain information in the format best suited to their current work.

[0160] The display unit 214 can provide different display formats depending on the type of business activity, as follows:

[0161] (1) In the case of product introduction work The display unit 214 can display product information in a visually easy-to-understand manner. For example, in the display format for product introduction tasks, a visually-centric layout can be adopted that makes extensive use of product images and thumbnail displays, allowing users to understand the features and advantages of a product at a glance. Search results are displayed categorized into categories such as "Basic Information," "Technical Specifications," "Case Studies," and "Comparison Tables," allowing users to quickly access the information they need. In addition, priority suggested products retrieved from the suggested product storage unit 232 are highlighted at the top of the screen to attract the user's attention.

[0162] Product introduction features include, for example, display functions such as product thumbnail gallery display, point display of features and benefits, horizontal scrolling display of related products, interactive display of competitor comparison tables, and presentation mode (customer-facing display).

[0163] (2) In the case of customer proposal work The display unit 214 can clearly show the relationship between customer challenges and solutions. For example, in the display format for customer proposal work, a category display by industry and challenge may be adopted, and knowledge tailored to the customer's attributes may be displayed preferentially. Search results may be classified into categories such as "proposal template," "case studies," "industry trends," and "solution overview," and materials for creating proposals may be systematically provided.

[0164] For customer proposal work, display functions such as category tree display (hierarchical structure of industry and challenges), proposal template preview display, display of similar customer case studies as cards, proposal structure support via drag and drop, and highlighting of proposal points are provided.

[0165] (3) In the case of sales management work: The display unit 214 can clearly show the progress of the sales process and the next steps. The display format for sales progress tasks centers around a chat interface and procedural guides to help users smoothly manage the sales process. Search results are organized according to process steps such as "contract procedures," "internal applications," "delivery plans," and "after-sales support," and information corresponding to the current progress can be highlighted. In addition, based on the progress information obtained from the progress detection unit 213, the next tasks to be performed and necessary documents can be recommended.

[0166] For sales operations, display functions such as process flow visualization, chat interface (question and answer format), checklist-style task management display, timeline display (progress history), and deadline / important date alerts may be provided.

[0167] The display unit 214 can provide display methods according to the content and format of the knowledge. For example, for PDF documents, a preview display and table of contents navigation may be automatically selected; for spreadsheet data, an interactive tabular display may be selected; and for presentation materials, a slideshow display may be selected. In particular, when displaying the results of a page-level search, only the relevant pages are extracted and displayed from a large amount of material, allowing the user to quickly access the information they need.

[0168] The display unit 214 can also optimize the display according to the user's environment and device. Different display formats are provided for each device, such as a detailed view that displays more information at once for desktop PCs, an interface optimized for touch operation for tablet devices, and a compact view that displays only the essential information for smartphones.

[0169] The display unit 214 can also be customized by the user. Users can customize the display layout, change the priority of display items, select a color theme, and more. These customization settings are stored in the user information storage unit 235, providing a display environment optimized for each user.

[0170] Furthermore, the display unit 214 can work in conjunction with the recommendation generation unit 218 to display related recommended knowledge in addition to search results. For example, recommendation information such as "Knowledge to See Also," "Frequently Used Knowledge," and "New Knowledge" can be displayed around the main search results to assist the user in their information search.

[0171] The display unit 214 can also provide operation guidance and help information. In particular, when using a function for the first time or when complex operations are required, context-appropriate hints and tutorials may be displayed.

[0172] A key feature of the display unit 214 is its three-tiered navigation structure: "For what purpose (sales operations)," "What (sales knowledge)," and "How (search method)." For example, the top of the screen can display a tab for selecting sales operations, the middle section can display sales knowledge categories, and the bottom section can display search method options in a hierarchical structure. Through this hierarchical structure, users can access the information best suited to their purpose.

[0173] <Input reception unit 215> The input receiving unit 215 is a processing unit that receives various inputs from the user and transmits them to the appropriate processing unit within the system. In the second embodiment, the input receiving unit 215 is configured as a software module implemented in the server and client terminals of the information processing system.

[0174] The input reception unit 215 provides a variety of input interfaces optimized according to the characteristics of sales operations. For example, it supports the following input formats:

[0175] (1) Free word search input The input reception unit 215 can accept keywords and phrases entered by the user as search queries. By entering keywords in the search bar at the top of the screen, users can search for relevant knowledge. During input, an input completion function is activated, and suggestions may be displayed for the keywords being entered. These suggestions may be generated based on the index information in the knowledge storage unit 231 and past search keywords in the search history storage unit 237. Synonyms and related keywords are also suggested to help the user select appropriate search terms.

[0176] (2) Category selection input The input reception unit 215 can accept input by selecting from a hierarchical category. By selecting categories such as industry (manufacturing, finance, service, etc.) or issue (business efficiency improvement, cost reduction, sales expansion, etc.), users can narrow down the relevant knowledge. Category selection is provided in the form of checkboxes, drop-down menus, tag clouds, etc., allowing users to operate intuitively. In particular, this category selection input is the primary input method in customer proposal work.

[0177] (3) Chat-style input The input reception unit 215 can accept questions and instructions in natural language in a chat format. By entering messages into the chat interface, interactive information retrieval with the system becomes possible. The entered text is analyzed using natural language processing, and appropriate searches and operations can be performed based on the user's intent. For example, in response to the input "Please show me a contract template," a search for contract templates will be performed. In particular, this chat-style input becomes the primary input method in sales management operations.

[0178] (4) Filter condition input The input receiving unit 215 can accept the specification of detailed filter conditions for the search results. By specifying conditions such as date range, creator, file format, and update frequency, the search results can be narrowed down more precisely. Filter conditions can be specified through interactive controls such as sliders, checkboxes, and date pickers.

[0179] (5) Input of operation commands The input reception unit 215 can accept operation commands such as button clicks, menu selections, and drag-and-drop. Through various on-screen operation elements, it can instruct users to perform operations such as search execution, knowledge viewing, saving, and sharing. These operation commands are optimized according to the type of sales work, and frequently used operations can be prioritized.

[0180] The input reception unit 215 also has a function to dynamically optimize the input interface according to the business operations identified by the business identification unit 211.

[0181] (1) In the case of product introduction work The input reception unit 215 can prioritize providing free-word searches related to product names and features. It may also display input elements specifically for product information searches, such as quick selection buttons for product categories and checklists of key product comparison items. Filters that allow users to specify product technical specifications and price ranges using sliders may also be provided.

[0182] (2) In the case of customer proposal work The input reception unit 215 can prioritize the selection of industry and issue categories. It can also display input elements specific to customer proposals, such as a quick input form for customer attributes and buttons for selecting the proposal phase. For example, a proposal template selection interface is provided, allowing users to quickly specify a template suited to their needs.

[0183] (3) In the case of sales management work The input reception unit 215 can prioritize providing natural language input in a chat format. It can also display input elements specific to sales progress, such as steps in the sales process and quick-selection buttons for frequently asked questions. For example, a checklist input interface for sales tasks is provided, allowing for quick input of progress. In particular, the "contract document space" is set as the default search target, making it easy to access knowledge related to contracts and deliveries.

[0184] In the chat-style input, a function is provided to display frequently used questions and instructions as quick-selection buttons by default. Relevant questions and instructions are automatically displayed according to the user's work context. For example, in "sales progress" tasks, "Show me a contract template," "Tell me the delivery schedule," and "What is the billing procedure?" are displayed. These default questions are dynamically generated and updated based on past search trends recorded in the search history storage unit 237 and the current sales tasks identified by the task identification unit 211. Users can complete their questions in natural language simply by clicking the displayed question buttons, enabling efficient information access.

[0185] The input reception unit 215 also has a function to personalize the input interface based on the user's usage patterns. Based on the user's attributes (specialty field, industry of expertise, years of experience, etc.) stored in the user information storage unit 235 and past input patterns recorded in the search history storage unit 237, it can prioritize the display of frequently used input elements and suggest frequently used keywords. It can also prioritize the display of spaces that the user has frequently used in the past.

[0186] The input reception unit 215 can provide hierarchical input navigation based on a three-tiered structure of "For what purpose (sales operations)", "What (sales knowledge)", and "How (search method)". Users can efficiently and purposefully input data by first selecting the target sales operations, then specifying the type of sales knowledge required, and finally entering specific search conditions.

[0187] <Knowledge Classification Section 217> The knowledge classification unit 217 is a processing unit responsible for systematically classifying the diverse knowledge stored in the knowledge storage unit 231, enabling efficient searching and utilization. The knowledge classification unit 217 can systematically classify knowledge from multiple perspectives. Possible classification axes include the following:

[0188] (1) Business classification The Knowledge Classification Unit 217 can classify each piece of knowledge based on the sales operations (product introduction, customer proposal, sales progress, etc.) defined in the Sales Operations Memory Unit 236. For example, product catalogs and data sheets are classified under "product introduction" operations, proposal templates and case studies under "customer proposal" operations, and contract templates and procedural guides under "sales progress" operations. This operation-based classification allows for the efficient extraction of knowledge related to the sales operations identified by the Operation Identification Unit 211.

[0189] (2) Knowledge Type Classification The Knowledge Classification Unit 217 can classify knowledge based on its type and format. Major knowledge types include "product information," "technical information," "proposal information," "case studies," "contract information," "manuals," and "FAQs." Each type is further subdivided into more detailed subcategories. For example, "product information" is subdivided into "catalogs," "datasheets," and "price lists."

[0190] (3) Classification by customer attributes The knowledge classification unit 217 can classify knowledge based on customer attributes. Key customer attributes include "industry," "company size," "position," and "challenges." For example, knowledge can be classified into categories such as case studies for manufacturing companies, proposals for small and medium-sized enterprises, presentation materials for management, and solution materials for cost reduction challenges. This classification can be used to provide appropriate knowledge based on customer information obtained from the customer information acquisition unit 216.

[0191] (4) Classification by content characteristics The knowledge classification unit 217 can classify knowledge based on its content and characteristics. For example, it can classify knowledge from perspectives such as "basic information," "detailed specifications," "comparative information," "success stories," "implementation procedures," and "FAQ." This classification makes it possible to provide knowledge that matches the type and level of detail that users require.

[0192] The knowledge classification unit 217 can automatically perform classification processing when new knowledge is registered or when existing knowledge is updated. Alternatively, the classification information for all knowledge may be updated through periodic batch processing.

[0193] The knowledge classification unit 217 can automatically analyze the content of knowledge and determine the appropriate category by utilizing technologies such as text analysis, morphological analysis, and machine learning. For example, it can extract characteristic keywords from the text content of the knowledge and determine business categories and customer industries based on the occurrence patterns of those keywords. To improve the accuracy of automatic classification, a supervised learning approach can also be adopted, utilizing past manual classification results as training data.

[0194] The knowledge classification unit 217 can also provide an automatic classification tag suggestion function. When registering or updating knowledge, it suggests appropriate tags based on content analysis, allowing users to easily assign classification information. For example, a list of suggested tags is displayed, and the user can select or edit the appropriate tag.

[0195] The knowledge classification unit 217 can also manage and optimize the classification system itself. It continuously improves the category structure and classification criteria based on an analysis of knowledge usage and search patterns. For example, if a particular category becomes excessively large, it automatically suggests more subdivided subcategories. Furthermore, categories that are rarely used may be suggested for consolidation or elimination.

[0196] The knowledge classification unit 217 can also provide a function for analyzing the relationships between knowledge items. Based on similarity of content, reference relationships, co-occurrence patterns, etc., it quantifies the degree of relevance between knowledge items and presents them as related knowledge. This relevance information can also be used for knowledge recommendations by the recommendation generation unit 218.

[0197] For example, in category searches, hierarchical navigation can be provided based on the classification system constructed by the Knowledge Classification Unit 217. Users can start with business categories and then gradually narrow down their search by knowledge type, customer attributes, and so on.

[0198] The knowledge classification unit 217 can also support the classification and display of search results. By grouping and displaying results from free-word searches and other searches by category, it enables users to efficiently find the information they need. For example, the search results for "cloud ERP" can be displayed divided into categories such as "product overview," "technical specifications," "implementation examples," and "pricing information."

[0199] The classification information from the knowledge classification unit 217 is also used to customize the user interface by the display unit 214. Categories related to sales operations identified by the business identification unit 211 are displayed preferentially, allowing users to quickly access the knowledge necessary for their current work.

[0200] <Recommendation generation unit 218> The recommendation generation unit 218 recommends the most suitable knowledge based on the user's current sales activities and customer context. By utilizing multiple information sources and analytical algorithms, the recommendation generation unit 218 can recommend the most suitable knowledge to the user through approaches such as the following:

[0201] (1) Business context-based recommendations The recommendation generation unit 218 can recommend knowledge that is particularly useful in a given business phase based on the current sales business information obtained from the business identification unit 211 and the progress status obtained from the progress detection unit 213. For example, in the initial stages of a "product introduction" task, product overview materials and comparison tables may be recommended, while in the middle stages, technical specifications and implementation examples may be recommended. In a "customer proposal" task, proposal templates and success stories may be preferentially recommended. In a "sales progress" task, contract templates and procedural guides may be recommended.

[0202] (2) Customer attribute-based recommendations The recommendation generation unit 218 can recommend knowledge that is suitable for the customer's attributes based on the customer information (industry, size, challenges, etc.) obtained from the customer information acquisition unit 216. For example, a manufacturing customer might be recommended case studies and proposal materials for the manufacturing industry, while a customer whose challenge is cost reduction might be recommended materials on solutions with high cost-reduction potential.

[0203] (3) Recommendations based on suggested products The recommendation generation unit 218 can recommend knowledge related to a product based on the priority product information obtained from the suggested product storage unit 232. For example, if "Cloud ERP" is the priority product, its catalog, technical specifications, case studies, and comparison tables may be recommended preferentially.

[0204] (4) Collaborative filtering-based recommendations The recommendation generation unit 218 can recommend knowledge that is frequently used in similar situations based on the behavior patterns of other users recorded in the search history storage unit 237. It extracts patterns such as "knowledge frequently used by other sales representatives for this task / customer" and "knowledge that users who viewed this knowledge often view next" and utilizes them for recommendations.

[0205] (5) Content-based recommendations The recommendation generation unit 218 can recommend knowledge with similar content to knowledge recently viewed by the user, based on classification information and content analysis by the knowledge classification unit 217. Related knowledge can be identified using indicators such as text similarity, common keywords, and same category.

[0206] The recommendation generation unit 218 can employ a weighting method that considers multiple factors in calculating the recommendation score. For each piece of knowledge, it can calculate scores for each element such as business relevance, customer suitability, suggested product relevance, collaboration score, and content similarity, and then integrate these scores with appropriate weights to determine the final recommendation score. The weighting parameters can be continuously optimized by machine learning based on the usage and results of past recommendation results.

[0207] The recommendation generation unit 218 can also optimize the type and display format of recommendations according to the situation. The main recommendation display formats are as follows:

[0208] (1) Top Recommendations This format highlights the most relevant knowledge in a prominent position on the screen. For example, it may appear at the top of the search screen as "Recommended Knowledge." In particular, knowledge directly related to the business identified by the Business Identification Unit 211, or knowledge related to high-priority products in the Proposed Product Storage Unit 232, may be displayed in this format.

[0209] (2) Related Knowledge Recommendations This format recommends other knowledge related to the knowledge currently being viewed or displayed as a search result. For example, while viewing a product catalog, technical specifications and case studies may be displayed alongside it as "Related Knowledge." Recommendations based on content similarity or collaborative filtering can be displayed in this format.

[0210] (3) Contextual recommendation This format proactively recommends knowledge that is likely to be needed next, based on current sales operations and progress. For example, during the customer proposal preparation phase, proposal templates and price lists might be displayed as "knowledge commonly used at this stage." Context-based recommendations can be displayed in this format.

[0211] (4) Personalized recommendations This format recommends personalized knowledge based on user attributes and past usage patterns. For example, it may display as "Knowledge that matches you well" or "Materials related to recently viewed knowledge." Recommendations based on user information and content-based recommendations can be displayed in this format.

[0212] The recommendation generation unit 218 can have functions for measuring and optimizing the effectiveness of recommendations. It tracks whether the recommended knowledge is actually viewed and used by the user and learns highly effective recommendation patterns. It also has a function to compare and verify the effectiveness of different recommendation algorithms using an A / B testing approach and select the more effective method.

[0213] The recommendation generation unit 218 can also provide search query extension capabilities. By automatically adding related keywords and synonyms to the search keywords entered by the user, it improves the accuracy and comprehensiveness of search results. For example, if the keyword is "CRM," the search may include synonyms such as "customer relationship management" and "customer relationship management."

[0214] The recommendation generation unit 218 can detect changes in sales status in real time based on user operations and information from the progress detection unit 213, and dynamically update the recommendation content. For example, if a new issue is identified during a conversation with a customer, knowledge related to that issue will be immediately recommended.

[0215] <Operation> Figure 8 is a diagram illustrating the operation of the management server 2.

[0216] The management server 2 uses a business identification unit 211 to identify the user's sales business (S301), selects a search method appropriate to the sales business (S302), and searches for knowledge from the knowledge storage unit 231 (S303). The management server 2 also selects relevant knowledge according to the user's situation (S304) and displays it in a format suitable for the sales business (S305).

[0217] In the operation shown in Figure 8, in S302, the management server 2 not only selects a search method according to the sales task, but also sets optimal search conditions as default values. Specifically, based on the sales task identified by the task identification unit 211, the search unit 212 refers to the task-specific search condition table 233d and sets search conditions (search keywords, filter conditions, target space, etc.) suitable for that task. Users can modify these default search conditions as needed. For example, for the "product introduction" task, search conditions focused on product-related keywords and the latest product information are set, while for the "customer proposal" task, search conditions related to the customer's industry and challenges are set.

[0218] As described above, the information processing system of the second embodiment can automatically identify the sales tasks performed by the user (product introduction, customer proposal, sales progress, etc.) and search for knowledge using the search method best suited to those tasks. Furthermore, by recommending the optimal space according to the sales task and setting appropriate search conditions as default values, sales representatives can efficiently access appropriate information for each sales scenario. In addition, users can change the recommended space and search conditions as needed, enabling flexible information retrieval.

[0219] The second embodiment has been described above, but the above embodiment is for the purpose of facilitating understanding of the present invention and is not intended to limit the interpretation of the present invention. The present invention can be modified and improved without departing from its spirit, and equivalents thereof are also included in the present invention.

[0220] For example, the processing performed by each functional unit of the management server 2 described above may be executed by any of the functional units. Furthermore, different functional units may be added to perform some of the processing performed by each of the functional units described above. Also, the functional units of the management server 2 may be distributed across multiple computers.

[0221] Furthermore, the information stored in each memory unit of the management server 2 may be stored in any of the memory units. That is, the information stored in the multiple memory units mentioned above may be stored in a single memory unit, or a portion of the information stored in one memory unit may be stored in another memory unit.

[0222] <Variation 10> In the second embodiment, relatively abstract sales activities such as "product introduction," "customer proposal," and "sales progress" were described as "business operations," but these can also be interpreted as higher-level concepts of more specific business activities. In this modified example, these abstract business operations are called "business sections" and are distinguished from more specific business tasks.

[0223] A "business section" is a higher-level concept representing the main stages or phases of sales activities, and each section consists of multiple specific business tasks. For example, a "product introduction" business section includes specific business tasks such as "product overview explanation," "detailed function explanation," "competitor comparison," and "price explanation." Similarly, a "customer proposal" business section includes business tasks such as "problem assessment," "solution proposal," "ROI explanation," and "implementation plan presentation," while a "sales progress" business section includes business tasks such as "contract terms negotiation," "internal approval," "delivery schedule adjustment," and "after-sales support plan."

[0224] By introducing this hierarchical structure, it becomes possible to achieve optimal knowledge retrieval that is more granular and tailored to the specific business context. The business identification unit 211 can identify not only business sections but also more specific business tasks based on the user's behavior patterns and operations. This allows the search unit 212 to select search methods and search conditions that are more suitable for the specific business tasks.

[0225] For example, even within a business section like "Product Introduction," a task like "Product Overview Description" prioritizes summary documents and feature lists, while a task like "Competitor Comparison" prioritizes comparison tables and differentiating points documents. This allows for search optimization tailored to more specific business contexts. This optimization is achieved through the following specific processes for each business task:

[0226] (1) Adjusting the weighting of search target fields: In "Product Overview," set high weights on the "Features" and "Benefits" fields, and in "Competitor Comparison," set high weights on the "Differentiation Points" and "Comparison Items" fields.

[0227] (2) Changes to the search result ranking algorithm: For "product description," a ranking that emphasizes conciseness and visual descriptiveness will be applied, while for "competitor comparison," a ranking that emphasizes comprehensiveness and quantitative comparative data will be applied.

[0228] (3) Default search filter settings: For "Product Overview," the "Overview Document" and "Catalog" categories are set as the initial filters. For "Competitor Comparison," the "Comparison Table" and "Market Analysis" categories are set as the initial filters.

[0229] Furthermore, even within the "Customer Proposal" business section, the optimal search criteria and target space are set for each business task. For example, the "Issue Hearing" task prioritizes industry-specific issue lists and interview sheets, while the "ROI Explanation" task prioritizes ROI calculation sheets and implementation effect case studies.

[0230] The optimization parameters for each business task are stored in the business task optimization table 233e within the search method storage unit 233. This table includes fields such as business task ID, search method ID, parameter type, parameter value, and application conditions. For example, the following parameters are set for the "Product Overview Description" task: • Search keyword expansion: "Overview" → "Overview, Features, Basic Functions, Benefits of Implementation" • Search target space: Narrow down to "product catalog space" • Sort search results by "Visual Clarity Score" • Display element priority: Prioritize knowledge containing "images," "charts," and "bullet points."

[0231] To accommodate this modification, the structure of the sales operations storage unit 236 may also be extended. A business section flag is added to the sales operations master table 236a, allowing identification of whether each record is a business section or a specific business task. Additionally, a business hierarchy table 236e is newly added to define the parent-child relationships between business sections and business tasks. The business hierarchy table 236e includes fields such as parent business ID, child business ID, relationship type, and order.

[0232] The business identification unit 211 can utilize this hierarchical structure to take a two-step approach: first, identification at the business section level, and then, if possible, identification of more specific business tasks. If specific business tasks are identified, search methods and conditions optimized for those tasks are applied; if identification at the business task level is difficult, optimization based on higher-level business sections is performed. This two-step approach is implemented as follows:

[0233] (1) Business section specific processing: • Generate feature vectors from user operation patterns (selected menus, previous operations, currently used knowledge, etc.) • Input feature vectors into the business section identification model and calculate the probability score for each business section. • Identify the business section with the highest probability score as the first candidate. If the probability score exceeds a threshold (e.g., 0.7), it is confirmed; otherwise, it is put on hold.

[0234] (2) Business task identification process: • Narrow down the candidates to only the business tasks belonging to the specified business section. • Analyze task-specific characteristics (keyword occurrences, operation sequences, document access patterns, etc.) • Input into the business task identification model and calculate the probability score for each business task. • Identify the task with the highest probability score as the primary candidate. • If the probability score exceeds a threshold (e.g., 0.6), the result is confirmed; otherwise, optimization at the business section level is applied.

[0235] Once a specific business task is identified, the search unit 212 reads the optimization parameters specific to that task from the business task optimization table 233e and applies them to the search process. This application is performed in the following steps:

[0236] • Select search method: Choose the search method best suited to your task (e.g., for the "Competitor Comparison" task, select "Table Search"). · Setting of search parameters: Set the parameters of the selected search method to the task-optimal values · Query expansion and transformation: Expand and transform the user's input query based on the task-optimization rules · Filtering of search results: Apply advanced filtering based on task relevance · Ranking adjustment: Re-rank the search results based on task-specific importance metrics

[0237] In identifying business tasks and adjusting optimization parameters, machine learning techniques are also utilized. Specifically, the following approaches are adopted: · Supervised learning: Model the relationship between business tasks and user behavior from past usage histories · Reinforcement learning: Continuously adjust the optimization parameters using the usage status of search results (such as browsing, utilization, evaluation, etc.) as feedback · Clustering: Cluster similar search patterns to discover new business task candidates · Transfer learning: Learn from the optimization parameters of other users and apply them to new users and unknown tasks

[0238] Through these machine learning approaches, the system autonomously adjusts the optimization parameters according to the usage situation, achieving more effective knowledge search. For example, in a specific business task, if a certain parameter setting has received high evaluations from many users, the probability of applying that parameter value to other users increases. Also, personalized optimization parameters are provided according to the usage patterns of individual users.

[0239] <Disclosure> Note that the present disclosure also includes the following configurations. [Item 1] A knowledge storage unit that stores a plurality of knowledges, A task identification unit that identifies the business operations performed by the user, A search unit that searches the knowledge by different search methods according to the identified business operations, An information processing system characterized by comprising the above. [Item 2] The information processing system described in item 1, It is equipped with a suggested product storage unit that stores products to be suggested to the customer, The search unit, with respect to the sales activities of product introduction, identifies the products to be proposed by referring to the proposed product storage unit, and searches for the knowledge related to the identified products to be proposed. An information processing system characterized by the following. [Item 3] The information processing system described in item 1, The knowledge storage unit stores, in association with the knowledge, at least one of the customer's industry or the customer's challenges. The search unit receives a designation of at least one of the customer's industry or the customer's challenges with respect to the sales operations of the customer proposal, and searches for the knowledge corresponding to at least one of the customer's industry or the customer's challenges that was received. An information processing system characterized by the following. [Item 4] The information processing system described in item 1, The system includes a search method storage unit that stores information that identifies the search method in correspondence with the aforementioned sales operations, The search unit identifies the search method to be used by referring to the search method storage unit. An information processing system characterized by the following. [Item 5] The information processing system described in item 1, It is equipped with a progress detection unit that detects the progress of the user's sales activities, The aforementioned business identification unit estimates sales operations based on the progress status, An information processing system characterized by the following.

[0240] <Variation 11> In the embodiment described above, the attribute value receiving unit 213 receives attribute value specifications from the user, but the knowledge data may also be searched using a question-and-answer format.

[0241] The management server 2 may include a question storage unit that stores information identifying a question, information identifying candidate answers to that question, and information identifying attributes and attribute values ​​corresponding to those candidate answers, in association with a business process. The question storage unit can store information indicating the order of multiple questions.

[0242] When the task designation unit 211 receives a task designation from the user, the attribute identification unit 212 can refer to the question storage unit to identify the question corresponding to the received task. The attribute identification unit 212 can select the first question from the identified questions and create screen data to send to the user terminal 1, along with the answer candidates corresponding to that question.

[0243] The attribute value receiving unit 213 receives the user's selection of an answer. The attribute value receiving unit 213 can refer to the question storage unit to identify the attribute and attribute value corresponding to the received answer.

[0244] The attribute identification unit 212 can refer to the question storage unit to identify questions in the following order and create screen data to send to the user terminal 1 along with candidate answers corresponding to those questions. By repeating the question and answer process in this way, attribute values ​​can be received sequentially.

[0245] The search unit 214 can search the knowledge data using the attribute values ​​corresponding to all the responses received by the attribute value receiving unit 213 as search conditions.

[0246] In this way, users can easily search for knowledge data simply by answering pre-set questions.

[0247] <Variation 12> In the above embodiment, the order of questions stored in the question storage unit was described as fixed, but the next question may be dynamically changed according to the user's answer.

[0248] The question memory unit can store information for specifying a question, information for specifying answer candidates, and information for specifying the next question to be presented for each answer candidate, in association with the business. Thereby, the branching logic of the question can be realized.

[0249] For example, for the first question "What is the customer scale?", the question memory unit can store the answer candidates "large enterprise" and "small and medium-sized enterprise", and for the answer candidate "large enterprise", store "What is the industry?" as the next question, and for the answer candidate "small and medium-sized enterprise", store "What is the problem?" as the next question.

[0250] The attribute specifying unit 212 can dynamically select the next question to be presented by referring to the question memory unit according to the answer received by the attribute value receiving unit 213.

[0251] By doing so, it is possible to follow the optimal question path according to the user's answer content, and it is possible to search for knowledge data more efficiently.

[0252] <Modified Example 13> In the above-described embodiment, the attributes associated with the knowledge data have been described. However, the knowledge data may be classified by category, and the search target may be controlled in units of categories.

[0253] The knowledge memory unit 231 can store information (hereinafter referred to as category information) for specifying the category to which the knowledge data belongs, in association with the knowledge data. The category can be, for example, "product flyer", "specification document", "proposal document", "case collection", etc. One knowledge data can belong to multiple categories.

[0254] The job memory unit 233 can store information for specifying the category to be searched, in association with the business. For example, for the business of "product introduction", the categories of "product flyer" and "specification document" can be stored as the search target.

[0255] When the task designation unit 211 receives a task designation from a user, the attribute identification unit 212 can refer to the job storage unit 233 to identify the category corresponding to the accepted task.

[0256] The search unit 214 can search only for knowledge data belonging to a specified category and retrieve knowledge data corresponding to attribute values.

[0257] Furthermore, the management server 2 can display a category selection screen on the user terminal 1. In this case, the categories identified by the attribute identification unit 212 can be displayed in a default selected state (for example, with the checkbox checked). The user can use the default selected categories as they are, or they can change their selection.

[0258] The search unit 214 can search for knowledge data belonging to the category selected by the user on the category selection screen.

[0259] This approach allows for efficient searching of knowledge data in the appropriate category according to the task.

[0260] <Example 14> Management server 2 can provide an administration screen for operators to manage question sets.

[0261] The management server 2 may include a management unit that receives input from the operator terminal, including information that identifies the task, information that identifies the question, information that identifies the answer candidate, information that identifies the attribute and attribute value corresponding to the answer candidate, and information that identifies the next question corresponding to the answer candidate.

[0262] The management unit can store the received information in the question storage unit. The management unit can also receive input information that identifies the task and information that identifies the category to be searched, and store it in the job storage unit 233.

[0263] The management department can provide an interface that allows for visual editing of the question order and branching logic. For example, it can display the flow of questions and answers in a flowchart format, allowing users to add, delete, rearrange questions, and configure branching settings using drag-and-drop operations.

[0264] Furthermore, the management unit can provide an interface for adding, changing, and deleting categories. The management unit can receive input from the operator, including information identifying a category and information identifying the knowledge data belonging to that category, and store this information in the knowledge storage unit 231.

[0265] In this way, operators can easily manage the optimal question sets and category settings according to the characteristics of the work.

[0266] <Example 15> The question-and-answer format search and category filtering described in the above variations 11 to 14 may be integrated.

[0267] The question memory unit can store information that identifies categories in addition to, or instead of, information that identifies attributes and attribute values, in association with candidate answers.

[0268] When the attribute value receiving unit 213 receives a response from the user, the attribute identification unit 212 can refer to the question storage unit to identify the category corresponding to the received response.

[0269] The search unit 214 can identify multiple categories corresponding to multiple answers and narrow down the search target by logical AND or OR of the identified categories. For example, if the category corresponding to the first answer is "product flyer" and the category corresponding to the second answer is "for large corporations," the search target can be knowledge data belonging to the categories "product flyer" and "for large corporations."

[0270] Furthermore, the attribute identification unit 212 can also refer to the question storage unit to identify both the attribute and attribute value corresponding to the answer, as well as the category. In this case, the search unit 214 can perform a search by combining filtering by attribute value and filtering by category.

[0271] By doing so, information obtained through the question-and-answer format can be utilized from both attribute values ​​and categories to achieve more accurate knowledge searches. [Explanation of Symbols]

[0272] 1. User terminal 2 Management Server

Claims

1. An attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, A job storage unit that stores information identifying the aforementioned attributes related to a task, in correspondence with the task, The task assignment department accepts task assignments from users, An attribute identification unit that identifies the attributes corresponding to the received business by referring to the job storage unit, An attribute value receiving unit that receives the specification of the attribute value from the user for at least one of the identified attributes, A search unit that retrieves the knowledge data corresponding to the received attribute value from the attribute value storage unit, An information processing system characterized by comprising the following features.

2. The information processing system according to claim 1, The attribute value receiving unit retrieves the attribute value stored in the attribute value storage unit without duplication for each of the identified attributes, and outputs a list of the attribute values ​​for each attribute. An information processing system characterized by the following.

3. The information processing system according to claim 1, Regardless of the attribute, the attribute value receiving unit receives text data input from the user. The search unit searches for knowledge data that corresponds to the received attribute value and in which the received text data matches at least one of the contents of the knowledge data or the attribute value of the attribute associated with the knowledge data. An information processing system characterized by the following.

4. The information processing system according to claim 1, The job storage unit stores information that identifies the attributes in association with the business and industry, The attribute value receiving unit receives the designation of the business and industry from the user, and identifies the attribute corresponding to the received business and industry. An information processing system characterized by the following.

5. The information processing system according to claim 1, The system includes a user information storage unit that stores the type of industry in which each user is engaged. The job storage unit stores information that identifies the attributes in association with the business and industry, The attribute value receiving unit refers to the user information storage unit to identify the industry corresponding to the user, and to identify the business specified by the user and the attribute corresponding to the industry corresponding to the user. An information processing system characterized by the following.

6. The steps include storing attribute values ​​in an attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, The steps include: storing information in the job storage unit that identifies the attributes related to the business in accordance with the business; The steps include: referring to the job storage unit to identify the attributes corresponding to the task specified by the user; A step of receiving a specification of the attribute value from the user for at least one of the identified attributes, The steps include: searching the attribute value storage unit for the knowledge data corresponding to the received attribute value; An information processing method characterized by a computer executing the following.

7. The steps include storing attribute values ​​in an attribute value storage unit that stores attribute values ​​for each of the multiple attributes associated with the knowledge data, The steps include: storing information in the job storage unit that identifies the attributes related to the business in accordance with the business; The steps include: referring to the job storage unit to identify the attributes corresponding to the task specified by the user; A step of receiving a specification of the attribute value from the user for at least one of the identified attributes, The steps include: searching the attribute value storage unit for the knowledge data corresponding to the received attribute value; A program that causes a computer to execute something.

8. A knowledge storage unit that stores multiple pieces of knowledge, The Business Identification Department identifies the sales tasks performed by the user, A search unit that searches for the aforementioned knowledge using different search methods according to the identified sales operations, An information processing system characterized by comprising the following features.

9. The information processing system according to claim 8, It is equipped with a suggested product storage unit that stores products to be suggested to the customer, The search unit, with respect to the sales activities of product introduction, identifies the products to be proposed by referring to the proposed product storage unit, and searches for the knowledge related to the identified products to be proposed. An information processing system characterized by the following.

10. The information processing system according to claim 8, The knowledge storage unit stores, in association with the knowledge, at least one of the customer's industry or the customer's challenges. The search unit receives a designation of at least one of the customer's industry or the customer's challenges with respect to the sales operations of the customer proposal, and searches for the knowledge corresponding to at least one of the customer's industry or the customer's challenges that was received. An information processing system characterized by the following.

11. The information processing system according to claim 8, The system includes a search method storage unit that stores information that identifies the search method in correspondence with the aforementioned sales operations, The search unit identifies the search method to be used by referring to the search method storage unit. An information processing system characterized by the following.

12. The information processing system according to claim 8, It is equipped with a progress detection unit that detects the progress of the user's sales activities, The aforementioned business identification unit estimates sales operations based on the progress status, An information processing system characterized by the following.

Images