JP7408957B2 - Idea proposal support system, idea proposal support device, idea proposal support method and program - Google Patents

Description

The present invention relates to an idea proposal support system, an idea proposal support device, an idea proposal support method, and a program.

As a technology related to electronic computers, a technology for extracting data related to desired information from a large amount of data is known. There is also a technology (artificial intelligence: AI) that extracts rules based on large amounts of data, learns correspondence relationships, and performs pattern recognition and selects optimal solutions based on the learning results.

In contrast, humans are required to play a more creative role. However, it is difficult to evaluate the value of new ideas, and it often happens that valuable inventions go unimplemented or that effort is wasted on implementing low-value inventions. Patent Document 1 discloses a technique for evaluating the value of new ideas.

Japanese Patent Application Publication No. 2002-92229

However, in order to advance an idea for a new business or product within an organization to the implementation stage, it is generally necessary to create persuasive materials regarding the usefulness, feasibility, profitability, etc. of the idea. Gathering appropriate materials for new ideas and creating presentation materials during normal work requires a great deal of effort, and there is a problem in that useful ideas are often lost.

An object of the present invention is to provide an idea proposal support system, an idea proposal support device, an idea proposal support processing method, and a program that can effectively reduce the effort required to collect and generate materials related to idea proposals.

In order to achieve the above object, the invention according to claim 1:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output unit outputs integrated data integrated in a time series in association with the concept word.
This is an idea proposal support system that is characterized by the following.

The invention according to claim 2 is the idea proposal support system according to claim 1 ,
comprising a chart creation unit that creates chart data based on the search results related to the first type term;
The output unit is characterized in that it outputs the created chart data.

The invention according to claim 3 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
An idea characterized in that, when data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output unit outputs integrated data that is integrated with respect to the characteristic. It is a proposal support system .

The invention according to claim 4 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
The idea proposal support system is characterized in that the output unit combines a plurality of search results related to the second type term to generate and output combined data related to the characteristics.

The invention according to claim 5 is the idea proposal support system according to claim 3 or 4 ,
comprising a graph creation unit that creates a graph based on the search results related to the numerical scale,
The output unit is characterized in that it outputs the created graph.

The invention according to claim 6 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The idea proposal support system is characterized in that the search unit determines a priority order of the auxiliary terms to be combined with the concept word based on past search history information.

The invention according to claim 7 is the idea proposal support system according to claim 6 ,
The search unit is characterized in that it excludes concept words included in the search history information from the search terms.

The invention according to claim 8 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The idea proposal support system is characterized in that the search unit determines a priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term.

The invention according to claim 9 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
a determination unit that determines the relationship between the term included in the search result and the concept word ;
Equipped with
The idea proposal support system is characterized in that the search unit performs a re-search by adding a term determined to be highly relevant by the determination unit to the search term.

The invention according to claim 10 is,
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit is an idea proposal support system characterized in that, among the terms included in the search results, terms that satisfy conditions related to appearance frequency are added to the search conditions and the search is performed again.

The invention according to claim 11 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
An operation reception department ,
Equipped with
The idea proposal support system is characterized in that the search unit performs a search by adding the term accepted by the operation reception unit to the search term.

The invention according to claim 12 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
Non-text data including at least one of image data and audio data and text data can be input to the acquisition unit,
The idea proposal support system is characterized in that the acquisition unit recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.

The invention according to claim 13 is the idea proposal support system according to claim 12 ,
The non-text data includes image data,
The acquisition unit is characterized by having a trained model that performs recognition processing on the input image data and outputs text data indicating the content of the image data.

The invention according to claim 14 is the idea proposal support system according to claim 12 or 13 ,
The acquisition unit is characterized in that it extracts independent words from the input text data, and identifies and acquires the concept words according to the frequency of appearance of the independent words.

The invention according to claim 15 is the idea proposal support system according to any one of claims 12 to 14 ,
The acquisition unit is characterized by having a trained model that outputs the concept word of the text data in response to the input of the text data.

The invention according to claim 16 is :
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The acquisition unit identifies a related term to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector corresponding to the concept word. The idea proposal support system is characterized in that a text object specified by a meaning vector having a direction that satisfies a predetermined criterion with respect to the meaning vector is used as the related term.

The invention according to claim 17 is the idea proposal support system according to any one of claims 3 to 15,
The auxiliary terms include a type 1 term related to time.
The invention according to claim 18 is the idea proposal support system according to any one of claims 1, 2, and 17,
The first type term is characterized in that it includes at least a part of current events, circumstances, history, trends, eras, and chronology.
The invention according to claim 19 is the idea proposal support system according to any one of claims 1, 2, 6 to 18,
The auxiliary terms include a second type term related to a numerical scale.
The invention according to claim 20 is the idea proposal support system according to any one of claims 3 to 5 and 19,
The second type term includes at least a part of share, sales, profit, and number of people.
The invention according to claim 21 is the idea proposal support system according to any one of claims 1 to 15 and 17 to 20,
The acquisition unit is characterized in that it specifies a term related to the acquired concept word and adds the specified related term to the search terms.
The invention according to claim 22 is the idea proposal support system according to any one of claims 1 to 21 ,
The present invention is characterized by comprising a storage unit that stores the list of auxiliary terms.

The invention according to claim 23 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output unit outputs integrated data integrated in a time series in association with the concept word.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 24 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
When data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output unit outputs integrated data that is integrated with respect to the characteristic.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 25 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
The output unit combines a plurality of search results related to the second type term to generate and output combined data related to the characteristic.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 26 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit determines the priority order of the auxiliary terms to be combined with the concept word based on past search history information.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 27 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit determines the priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 28 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
a determination unit that determines the relationship between the term included in the search result and the concept word;
Equipped with
The search unit adds the term determined to be highly relevant by the determination unit to the search term and performs the search again.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 29 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit performs the search again by adding to the search conditions a term that satisfies a condition regarding frequency of appearance among the terms included in the search results.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 30 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
An operation reception department,
Equipped with
The search unit performs a search by adding the term accepted by the operation reception unit to the search term.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 31 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
Non-text data including at least one of image data and audio data and text data can be input to the acquisition unit,
The acquisition unit recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.
This is an idea proposal support device that is characterized by the following.
The invention according to claim 32 is:
an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The acquisition unit identifies a related term to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector corresponding to the concept word. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
This is an idea proposal support device that is characterized by the following.

The invention according to claim 33 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a type 1 term related to time,
In the output step, if data at a plurality of different timings are obtained in the search results related to the type 1 term, the integrated data integrated within the time series is output in association with the concept word.
It is characterized by
The invention according to claim 34 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a second type term related to numerical scale,
In the output step, if data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, output integrated data that is integrated about the characteristic.
It is characterized by
The invention according to claim 35 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a second type term related to numerical scale,
In the output step, a plurality of search results related to the second type term are combined to generate and output combined data related to the characteristic.
It is characterized by
The invention according to claim 36 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, the priority order of the auxiliary terms to be combined with the concept word is determined based on past search history information.
It is characterized by
The invention according to claim 37 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, a priority order of the auxiliary terms to be combined with the concept word is determined based on the relationship between the concept word and the auxiliary term.
It is characterized by
The invention according to claim 38 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
a determination step of determining the relationship between the term included in the search result and the concept word;
including;
In the search step, the term determined to be highly relevant in the determination step is added to the search term and the search is performed again.
It is characterized by
The invention according to claim 39 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, a search is performed again by adding to the search condition a term that satisfies the condition regarding the frequency of appearance among the terms included in the search results.
It is characterized by
The invention according to claim 40 is:
An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, a search is performed by adding the term accepted by the operation reception unit to the search term.
It is characterized by
The invention according to claim 41 is:
An idea proposal support method executed by a computer control unit,
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the acquisition step, non-text data including at least one of image data and audio data and text data can be input,
In the acquisition step, the content of the non-text data is recognized and converted into text data, and the concept word is acquired from the text data.
It is characterized by
The invention according to claim 42 is:
An idea proposal support method executed by a computer control unit,
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the acquisition step, a term related to the acquired concept word is identified, the identified related term is added to the search term, a meaning vector having a direction according to the meaning of the object is generated, and a meaning vector corresponding to the concept word is generated. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
It is characterized by

The invention according to claim 43 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output means outputs integrated data integrated in a time series in association with the concept word.
This program is characterized by:
The invention according to claim 44 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The auxiliary term includes a second type term related to numerical scale,
When data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output means outputs integrated data that is integrated with respect to the characteristic.
This program is characterized by:
The invention according to claim 45 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The auxiliary term includes a second type term related to numerical scale,
The output means combines a plurality of search results related to the second type term to generate and output combined data related to the characteristic.
This program is characterized by:
The invention according to claim 46 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The search means determines the priority order of the auxiliary terms to be combined with the concept word based on past search history information.
This program is characterized by:
The invention according to claim 47 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The search means determines the priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term.
This program is characterized by:
The invention according to claim 48 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
determining means for determining the relevance between the terms included in the search results and the concept word;
function as
The search means adds the term determined to be highly relevant by the determination means to the search term and performs the search again.
This program is characterized by:
The invention according to claim 49 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The search means adds to the search condition a term that satisfies a condition regarding frequency of appearance among the terms included in the search result and performs the search again.
This program is characterized by:
The invention according to claim 50 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The search means performs a search by adding the term accepted by the operation reception unit to the search term.
This program is characterized by:
The invention according to claim 51 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
Non-text data including at least one of image data and audio data and text data can be input to the acquisition means,
The acquisition means recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.
This program is characterized by:
The invention according to claim 52 is:
computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The acquisition means identifies a term related to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector that corresponds to the concept word. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
This program is characterized by:

According to the present invention, it is possible to effectively reduce the effort required to collect and generate materials related to idea proposals.

FIG. 1 is a block diagram showing the functional configuration of the idea proposal support system according to the present embodiment. FIG. 2 is a diagram illustrating the flow of idea proposal support. It is a figure which shows the example of the acquisition procedure of a similar word. FIG. 3 is a diagram illustrating an example of setting search terms based on image data. It is a figure which shows the example of a combination of numerical data. FIG. 3 is a diagram showing an example of integrating numerical data. It is a chart showing an example of a chronology that is created. 3 is a flowchart showing a control procedure of idea proposal support processing.

Embodiments of the present invention will be described below based on the drawings.
FIG. 1 is a block diagram showing the functional configuration of the idea proposal support system 1 of this embodiment.
This idea proposal support system 1 includes a server device 10 (idea proposal support device), a database device 20, and a terminal device 30.

The server device 10 includes a control unit 11 (an acquisition unit, a search unit, an output unit, a chart creation unit, a graph creation unit, a determination unit), a storage unit 12, a communication unit 13, and the like. The control unit 11 is a processor that includes a CPU (Central Processing Unit), a RAM (Random Access Memory), etc., and performs various arithmetic processes to centrally control the operation of the server device 10.

The storage unit 12 stores a program 121 executed by the control unit 11 and setting data. Furthermore, the storage unit 12 temporarily stores various data input from the terminal device 30 and its processed data. The storage unit 12 includes a nonvolatile memory such as an HDD (Hard Disk Drive) and/or a flash memory. Furthermore, the storage unit 12 may include a RAM or the like for temporarily storing large data and processing it at high speed. The program 121 includes programs related to text analysis, image recognition, voice recognition, and idea proposal support processing described below. A trained model 1211 that has been appropriately designed and trained can be used for each process related to image recognition, extraction of keywords (concept words of ideas), and the like. The setting data also includes an auxiliary term list 122 that stores a large number of auxiliary terms, which will be described later.

The communication unit 13 controls transmission and reception of data with external devices according to a predetermined communication standard such as TCP/IP (Transmission Control Protocol/Internet Protocol). The communication unit 13 is connected to external devices via a network. The network includes the Internet, and allows access to private search servers and the like using HTTP (Hypertext Transfer Protocol). The communication unit 13 may have a terminal that can directly communicate with an external device (such as a peripheral device) via a USB (Universal Serial Bus) or the like.

The database device 20 has a storage section 21 . The storage unit 21 stores terminology data 211 that associates a large number of objects expressed as text, images, sounds, etc. with their identification data (including values of meaning vectors described later), and terminology data 211 that associates a large number of objects expressed by text, images, sounds, etc. The search results and idea proposal history data 212 (past search history information) are stored and held. Further, the database device 20 may include a control unit that controls reading and writing of the storage unit 21 and a communication unit that controls communication with external devices.

The terminal device 30 receives input from a user and outputs information to the user. The terminal device 30 includes a control section 31, a storage section 32, a display section 33, an operation reception section 34, a communication section 35, and the like. The control unit 31 is a processor that has a CPU, a RAM, etc., and controls the operation of the terminal device 30 in an integrated manner. The storage unit 32 includes a nonvolatile memory and stores various programs and data. The communication unit 35 controls data transmission and reception with external devices according to a predetermined communication standard.

The operation reception unit 34 accepts an input operation from the outside (user, etc.) and outputs it to the control unit 31 as an input signal. The operation reception unit 34 includes, but is not particularly limited to, a keyboard, a mouse, a touch sensor provided over the display screen of the display unit 33, and the like, for example. The display section 33 has a display screen, and causes the display screen to display contents according to control commands from the control section 31. Although the display screen is not particularly limited, for example, it is a liquid crystal display screen (LCD). Further, the display section 33 may include an LED lamp (Light Emitting Diode) or the like that indicates a specific state.

Next, the operation details of idea proposal support in the idea proposal support system 1 of this embodiment will be explained. The server device 10 uses various databases and search sites to search for current and past related information using concept words based on the content of the input idea (idea data). Search targets include, but are not limited to, numerical data, particularly time series data and regional comparison data. Numerical data obtained from the search results are integrated to the extent possible and graphed and/or charted.

FIG. 2 is a diagram illustrating the flow of idea proposal support.
The input data indicating the content of the idea may be a list of concept words, a catchphrase, a summary sentence, an image or audio, or a combination thereof (can be input). Non-text data such as image data and audio data is first converted to text data. In the case of an image (sound), the image (sound) is converted into a term or explanatory text (text data) according to the content using a well-known image recognition technology (speech recognition technology) (PR1). As a well-known image recognition technique, for example, a trained model 1211 of a machine learning model designed by a convolutional neural network (CNN) or the like may be used. If the input or converted text data is an explanatory sentence or the like and is a combination of a plurality of words, concept words are identified and extracted using well-known language processing technology (PR2). A learned model 1211 related to a machine learning model of a predetermined algorithm may be used for the language processing technology. Different algorithms for machine learning models may be used depending on the language. As concept words to be identified and extracted, nouns may be preferentially extracted, or independent words in general including adjectives and verbs may be extracted. If concept words (words, short phrases, etc.) have been input from the beginning, these may be used as is. The processes of PR1 and PR2 are included in the acquisition step (acquisition step, acquisition means).

Although it is not necessary that only one concept word is obtained, if a plurality of concept words are obtained, some of them (mainly one word at a time) are selected at the time of search and the search is executed multiple times. Also, at this time, the search is executed not only using concept words but also using search terms (search conditions) combined with auxiliary terms extracted from the auxiliary term list 122 (PR3). The auxiliary term indicates an item of characteristic of a concept word, and is a term for acquiring data (search results) related to a characteristic that can be particularly expressed numerically. Auxiliary terms, that is, characteristic items, mainly include terms (type 1 terms) that express time (time, date and time, period, chronology), such as current events, circumstances, history, history, chronology, chronology, Terms (type 2 terms) that express numerical values (numerical scale) related to trends, business, etc., such as share, sales, profit, revenue, number of people, number of users, number of purchasers, number of visitors, etc. Examples include the number of participants, the number of stores, the number of animals (heads, wings), the number of pieces (pieces, machines, books), and the number of cases (times, books, degrees). In addition to these, auxiliary terms may include numerical scale classifications, terms related to attributes, such as generation (interval, separate, etc. may be added; the same applies hereinafter), male and female, sex, age, region, etc. good. Among these, terms that are suitable for combination with the concept word type (organisms, goods, services, commerciality, etc.) are preferentially selected, and the combination with the concept word and its order are prioritized. It is determined. In other words, particle words that are combined with general terms such as ``number'', such as ``person'' for small pets, ``feathers'' for birds and rabbits, and ``head'' for large animals such as horses and cows, are determined in advance. It may be stored and referred to and selected before searching.

In addition to the concept words and auxiliary terms automatically selected in this way, even if terms (designated words) accepted by the user's operation reception unit 34 and sent to the server device 10 are incorporated into the search term. good.

A search is performed using the search terms in the determined combination order (PR4). For the search, a closed database such as an in-house database, that is, the history data 212 of the database device 20, a free search server on the Internet, a paid database provided by a private research company, etc. may be used. Among these, searches of in-house databases, etc. may be performed before other searches. If numerical data, etc. are detected in the search results of the internal database (results of past searches), we may preferentially perform an external search using the same search term, or conversely, search using the same search term. If there are search results for this idea, it may be reflected in determining the priority order, such as excluding the idea from searches (search terms) on the Internet as it is not a new idea. The determination as to whether to prioritize or exclude may be made depending on, for example, the update date and time of the numerical data.

The object to be searched may include not only text data but also image data and audio data. Image data and audio data may be converted into text using a well-known recognition technique in the same manner as described above. Furthermore, the text data includes structured documents. For example, when a set of search results are obtained from search terms for each combination of search terms up to a predetermined priority level or for those whose priority level is higher than a predetermined standard, all search results or each search term can be obtained. The contents of a predetermined number of search results arranged in sequence (in case of duplicates may be excluded) are further analyzed to search for included words, and the frequently occurring words, concept words, and the frequently occurring words are analyzed. Extract terms similar to the word (PR5). If there are not enough search results, or if there are too many and it is unclear whether the appropriate results have been extracted, add related frequently occurring words (similar words) to the search term (if there are insufficient results, use logical OR). , if there are too many results, logical product, etc.). Frequently appearing words are words that satisfy conditions (conditions related to frequency of appearance), such as words that appear in a predetermined order of frequency of appearance, or have a frequency that exceeds a predetermined ratio compared to the number of characters. It is defined as. In addition, if there are frequently occurring words that are significantly different from the set concept words (for example, abbreviations with different meanings, proper nouns of fiction works, titles of various works, etc.), the content containing the frequently occurring words will be changed. Additional settings (search condition settings) may be made to the search term as an exclusion word so as to exclude the search term. With these, the search term is reset and the search is performed again (in the case of only adding a search pattern, only the search related to the added concept word needs to be performed anew). The priority order of search terms including similar words may be determined again according to the degree of similarity, relevance, etc. of the similar words to the concept word. Note that if a single search yields sufficient results, there is no need to perform the search again. Processes PR3 to PR5 are included in the search step (search unit, search means).

On the other hand, from the search results, data indicating characteristics corresponding to the auxiliary term, particularly data containing numbers (numerical values), are extracted (PR6). Contents indicated by numbers included in multiple search results are different classifications or classifications related to the same characteristic (for example, different calendar years or ages corresponding to one characteristic indicated by an auxiliary term (type 2 term)) classification, regional classification, correspondence conditions with other characteristics, etc.), it is possible to integrate data from multiple search results with different timing, conditions, classification, etc. within a time series into one dataset. Good (PR7). The integrated numerical data is appropriately graphed (graph data is created) or charted (chart data is created) (PR8; graph creation section, diagram creation section), and then the characteristics corresponding to the concept words are created. The data is output as shown. The processes of steps PR6 and PR7 are included in the output step (output unit, output means).

Extraction of words similar to concept words (related terms; hereinafter referred to as similar words) for use in searches is performed objectively based on numerical criteria. To extract such similar words, a meaning vector is used here. Semantic vectors are a technology that expresses directions according to the meanings of terms (objects) using multidimensional vectors. The angular difference (distance) between the dimensional vectors corresponds to each other. The number of dimensions may be arbitrarily determined, and may be, for example, 100 to 1000 dimensions. The semantic vector for each term is determined using a machine learning model, and the axial properties of each dimension do not need to be expressed linguistically.

Word2vec is known as a technology for determining semantic vectors using machine learning. Examples of learning algorithms related to machine learning include, but are not limited to, the Continuous Skip-gram model or the CBOW (Continuous Bag-of-Words) model that is often used in Word2vec. Further, the hierarchy of the neural network used is not limited to two levels, and further, learning algorithms other than the neural network may be used. Correspondence data between a large number of words and meaning vectors obtained by machine learning using a large amount of text data, etc. is held in the storage unit 21 of the database device 20 as term data 211. Machine learning may be performed on the server device 10 by preparing sufficient learning data in advance, or a learned model generated by learning performed on an external device may be copied to the database device 20. Note that the extraction of similar words is not limited to the case where it is performed based on the results of the previous (first) search. The concept words may be separately selected before the first search based on the concept words first acquired from the idea data. As mentioned above, the priority order for searching the extracted similar words is based on the degree of similarity (relevance) with the concept word (in particular, words with a high degree of relevance are given priority and added to the search term). May be determined.

FIG. 3 is a diagram illustrating an example of a similar word acquisition procedure.
In the first example shown in FIG. 3A, first, when a concept word is acquired, the concept word is sent to the database device 20 (PR21) and a corresponding meaning vector is acquired (PR22). Next, this meaning vector and the set angle (reference angle) for the meaning vector are sent to the database device 20 (PR23), and the term ( Text object) is acquired as a similar word (related term) (PR24). The obtained similar words are added to the list of search terms, and the priority order of combinations with auxiliary terms is determined in the same way as other concept words (PR25).

In the second example shown in FIG. 3(b), frequently appearing words and concept words obtained from the search results are sent to the database device 20 (PR31), and corresponding meaning vectors are calculated (PR32). Using the inner product of the obtained semantic vectors (value obtained by dividing the inner product by the product of the lengths of both semantic vectors), a value corresponding to the angular difference between the two semantic vectors is determined (PR33). It is determined whether the obtained angle difference is within the standard angle (relevance), and frequently occurring words within the standard angle are added to the list of search terms as similar words, and used in combination with auxiliary terms in the same way as other concept words. The priority order is determined (PR34).

FIG. 4 is a diagram showing an example of setting search terms based on image data.
This image data includes a bicycle P1 (road bike) as an object, and suggests, for example, an activity such as cycling as a concept. Image recognition is performed on the image data, and bicycles, road bikes, etc. (depending on the background, cycling, cycling road, road race, etc. may be included) are extracted as concept words.

For these concept words, auxiliary terms such as history, chronology, trend, number of units sold, number of users, price, and share may be selected. Further, based on the meaning vector, related terms such as other vehicle types such as mountain bikes and cross bikes, road racing, touring, cycling, long ride, wheel riding, and binding pedals may be obtained. Auxiliary terms for the related terms may also be selected. Based on these, possible combinations are selected, such as a combination of a road bike and the number of units sold, a combination of a road race and the number of participants, etc., and a search is performed using these as search terms by prioritizing these combinations.

As a result of a search using such a combination of search terms, numerical data, changes depending on the age, trends, etc. that are particularly useful as objective data can be more preferably obtained.

However, numerical data, especially those published on the Internet, are not necessarily all comprehensively published in a format suitable for presentation. For example, it is not possible to know what types of bicycles are popular just by looking at data on the number of bicycles sold.

FIG. 5 is a diagram showing an example of combining numerical data.
For example, as shown in FIG. 5(a), statistical data on the sales volume of all bicycles including car models A to C (multiple categories) is obtained through a search, and statistical data on the sales volume of all bicycles including car models A to C (multiple categories) is obtained through a search. When statistical data on the sales share of bicycles is obtained, these multiple search results are combined (in this case, multiplied) to roughly calculate the sales volume of bicycles of type A, and the data related to the desired results (combined) are calculated. data) can be generated. The graph of FIG. 5(b) is generated based on the numerical values of vehicle type A, thereby clearly showing the growth in sales of vehicle type A.

Note that it is often difficult to automatically determine whether such a combination is necessary. Therefore, for example, the priority order of parameters (characteristic values) and possible combination patterns for each parameter may be stored in the storage unit 12 in advance, and combinations may be performed according to possible combination patterns based on the acquired data. It is also possible to judge whether or not it is possible. Further, supplementary information for determining the priority order may be obtained together with the concept words.

In addition to cases where numerical data is explicitly shown, statistical data may be shown as a numerical table as image data, or only a graph may be displayed. In these cases, character recognition processing, graph image discrimination, numerical processing, etc. may be performed on the image. In addition, since it is time-consuming to perform high-precision character recognition processing on all image data, we perform simple image recognition processing to determine whether or not it is a graph or chart. High-precision processing may be performed only on image data.

FIG. 6 is a diagram showing an example of integrating numerical data.
As shown in Figure 6(a), in addition to slightly older statistical data on sales of product P, if an article showing the recent situation as shown in Figure 6(b) is obtained, these figures are integrated. can do. Here, information from news articles from 2000, 2012, and 2018 (different timings) is added to the data from 2002 to 2010 obtained from FIG. 6(a). The integrated data may be graphed as shown in FIG. 6(c). In this case, it can be seen that the value of the news article is an approximation, that is, the integrated data contains values with different accuracy, but as a material related to closed and informal presentations such as in-house, the accuracy is particularly high. differences may not be considered. Alternatively, a numerical arrangement table may also be written. In addition, while data differs depending on the source, differences in data acquisition conditions and data reliability may result in the inclusion of values with essentially different meanings or questionable values. If there is a description, annotations, footnotes, etc. may be added.

On the other hand, chronological data may be useful even if it is not necessarily quantitative, so it may be generated as an enumeration as tabular data.

FIG. 7 is a chart showing an example of a created chronology.
Here, we have created a chronology by arranging events related to road bikes in order, obtained from various news sources. In this way, by tracking the history of events held, not just changes in numbers, it becomes easier to track the current situation and future path.

FIG. 8 is a flowchart showing a control procedure by the control unit 11 of the idea proposal support process executed by the control unit of the server device 10 of this embodiment. This idea proposal support process is started in response to a request input from the terminal device 30.

When the idea proposal support process is started, the control unit 11 (CPU) acquires idea data from the terminal device 30 (step S101). The control unit 11 determines whether the acquired idea data is image data (step S102). If it is determined that the image data is image data (“YES” in step S102), the control unit 11 applies image recognition processing using the learned model 1211 or the like to the image data to identify objects in the image. is converted into text and an explanatory text of the contents is generated (step S103). Then, the process of the control unit 11 moves to step S104. If it is determined that the data is not image data ("NO" in step S102), the process of the control unit 11 moves to step S104.

When the process proceeds to step S104, the control unit 11 performs text processing and extracts concept words (step S104). The processes of steps S101 to S104 constitute an acquisition step (acquisition means in the program) in the idea proposal support process of this embodiment.

The control unit 11 refers to the auxiliary term list 122 and extracts auxiliary terms that can correspond to each concept word. The control unit 11 determines the search priority order by prioritizing the search term patterns that are combinations of each concept word and auxiliary term (step S105). The control unit 11 performs a search using the search terms in the determined priority order (step S106). As described above, when using both the internal database and external search, the internal database may be searched first, and the priorities of search terms may be changed based on the results.

The control unit 11 determines whether necessary data has been acquired (step S107). If it is determined that the data has not been acquired (“NO” in step S107), the control unit 11 controls the text of the data obtained as the search result (which may include text converted from image data. Words (independent words) are extracted from (for structured documents, structural analysis may be performed), the frequency is calculated, and words that appear more frequently than a standard are extracted. Further, as described above, the control unit 11 uses, for example, a meaning vector to acquire words similar to the concept word from among the frequently appearing words (step S108). The control unit 11 adds similar words to the search terms (step S109). The control unit 11 may add the above-mentioned exclusion word in the search to the search conditions.

The control unit 11 transmits the current search term list to the terminal device 30 and displays it on the display unit 33 (step S110). The control unit 11 waits for a reply from the terminal device 30, and upon receiving the reply, determines whether an instruction to add a search term has been given by the operation reception unit 34 or the like (step S111). If it is determined that there is one ("YES" in step S111), the specified word for which addition has been instructed is added to the search terms (step S112). Then, the process of the control unit 11 returns to step S105. If it is determined that there is no additional instruction (“NO” in step S111), the process of the control unit 11 returns to step S105.
The processes of steps S105 to S112 constitute a search step (search means in the program) in the idea proposal support process of this embodiment.

If it is determined in step S107 that the necessary data has been acquired (“YES” in step S107), the control unit 11 extracts numerical data from the search results, and converts image data etc. into text data as necessary. (Step S121). The control unit 11 determines the content of the numerical data and integrates data with matching content or related data that can be linked (step S122). The control unit 11 creates a graph from the numerical data related to the scale and the like, and also creates a chart such as a chronology as necessary (step S123). The control unit 11 outputs the generated data to the terminal device 30 (step S124). The control unit 11 may output numerical data and various generated data to the database device 20 and add them to the history data 212. Then, the control unit 11 ends the idea proposal support process.
The processes of steps S121, S122, and S124 constitute an output step (output means in the program) in the idea proposal support process of this embodiment.

As described above, the idea proposal support system 1 of this embodiment includes the control unit 11, which functions as an acquisition unit to acquire concept words, and as a search unit to search for concept words and the characteristics of the concept words. A search is performed using a combination with an auxiliary term indicating an item as a search term, and an output unit outputs data indicating characteristics corresponding to the concept word based on the search results.
In this way, the idea proposal support system 1 acquires data related to the characteristics of concept words and outputs them all at once, so it is possible to collect materials useful for creating documents necessary for proposing (presenting) a user's idea, It can help reduce the time and effort required for generation.

Further, the auxiliary terms include type 1 terms related to time. In order to present objective information, it is important to grasp the flow of related information in chronological order, so by selecting corresponding search terms, you can collect the information necessary for presentation more reliably. I can do it.

Further, the first type term includes at least a part of current events, circumstances, history, trends, eras, and chronology. By particularly utilizing these frequently occurring time-related terms, it is possible to increase the probability of collecting necessary information.

Furthermore, as an output unit, when data at a plurality of different timings are obtained in the search results related to the first type term, the control unit 11 outputs the integrated data integrated within the time series in association with the concept word. do. That is, when a plurality of search results that occur one after another in a time series are obtained, they are integrated and output, which greatly reduces the user's effort in comparing and arranging the plurality of analysis results.

Further, the control unit 11 acts as a chart creation unit to create chart data based on the search results related to the first type term, and as an output unit outputs the created chart data. In this way, the system not only outputs data according to the format of the search results, but also outputs the data together in charts and tables, which reduces the user's effort to create charts and visually enhances the content of proposed ideas. It is possible to easily make presentations.

Further, the auxiliary terms include type 2 terms related to numerical scale. Since numerical data is effective as objective data, it is possible to selectively add to the search terms terms that can more reliably acquire such numerical data. This makes it possible to more reliably acquire and output materials that are more effective for presentations.

Further, the second type term includes at least a part of share, sales, profit, and number of people. By particularly utilizing frequently occurring terms related to numerical values, it is possible to increase the probability of collecting quantitative data and more effectively improve the persuasiveness of presentations based on output information.

Further, as an output unit, when data in a plurality of different categories related to the same characteristic is obtained in the search results related to the second type term, the control unit 11 outputs integrated data that is integrated with respect to the characteristic. In other words, if multiple search results result in multiple numerical values for the same parameter with different timing or conditions (categories), they will be integrated and output as a list, allowing the user to compare and match the multiple search results. It is possible to easily organize, confirm, and present valid data without having to do so.

Further, the control unit 11 serves as an output unit, which combines a plurality of search results related to the second type term to generate and output combined data related to a desired characteristic. Even if the search results are based on Type 2 terms, they may not necessarily be data that directly contributes to improving the persuasiveness of idea proposals. Even in such cases, it may be possible to reach the necessary information by combining multiple parameters, so by combining the multiple parameters (search results) and generating combined data, you can easily obtain more effective output. becomes possible. This makes it possible to improve the persuasiveness of idea proposals without much effort.

Further, the control unit 11, as a graph creation unit, creates a graph based on the search results related to the numerical scale, and as an output unit, outputs the created graph. In the case of arrays of numerical data, visualizing them in graphs is very effective in helping third parties understand the proposal in a short time. Therefore, when a data array that can be graphed is obtained, by generating and outputting a graph, it is possible to easily create materials for effective idea proposals while saving the user the trouble of creating graphs. can.

Further, the control unit 11, as a search unit, determines the priority order of auxiliary terms to be combined with the concept word based on the past history data 212. If search results are stored on an internal server, the effort and time required to connect to an external server can be easily reduced by reusing the data. Further, if it is known in advance that there is no data useful for idea proposal, unnecessary effort and time can be reduced by not accessing the site again.

Further, the control unit 11, as a search unit, excludes concept words included in the history data 212 from search terms. In other words, search efficiency may be improved by excluding content that is known to have already been used in a search based on internal company history, as it has already been considered by someone. However, since the combinations of a plurality of concept words may be different, in such a case, corresponding history data may be acquired and used as described above.

Further, the control unit 11, as a search unit, determines the priority order of auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term. When searching, the presence or absence of related terms added to the search term is often important, so by adding a process to select highly related terms, it is possible to more effectively retrieve highly needed data. I can do it.

In addition, the control unit 11 acts as a judgment unit to judge the relevance between the terms included in the search results and the concept words, and acts as a search unit to add the terms determined to be highly relevant to the search terms and repeat the search. Perform a search.
In this way, highly relevant terms are extracted from the search results and added to the search terms, so each search result can be reanalyzed after narrowing down the search results to more effective results based on the actual search. Therefore, it is possible to easily reach the necessary information without wasting time and effort in analyzing the contents.

Further, the control unit 11, as a search unit, performs a re-search by adding to the search conditions a term that satisfies the condition regarding the frequency of appearance among the terms included in the search results. That is, by adding terms with a high number of hits in a search to the search terms for re-search or setting them as excluded words, it is possible to analyze the search content more efficiently.

Further, in the idea proposal support system 1, the terminal device 30 includes an operation reception unit 34. The control unit 11 of the server device 10, as a search unit, performs a search by adding the term accepted by the operation reception unit 34 to the search term. Completely automatic processing does not necessarily result in optimal search terms that reflect the idea base and background, depending on the content of the concept word and its relevance to the genre, so it is possible for the user to add to the search terms. By doing so, it is possible to perform efficient searches and generate output data with some additional effort on the part of the user, and as a whole, it is possible to reduce the effort required by the user to create idea proposal materials.

Further, as data for acquiring a concept word, non-text data including at least one of image data and audio data, and text data can be input, and the control unit 11, as an acquisition unit, can input the content of the non-text data. Recognize and convert into text data, and obtain concept words from the text data. In other words, in the idea proposal support system 1, even if a concept is expressed in non-text data, by appropriately converting it into text data once, concept words can be appropriately obtained in the same way as when the concept is originally text data. Is possible. This allows idea data in various forms to be appropriately linked to idea proposal materials, reducing the user's effort. In particular, by acquiring an appropriate combination of concept words from image data with a large amount of information, it is possible to save the user the trouble of converting it into text.

Further, the non-text data includes image data, and the control unit 11, as an acquisition unit, performs recognition processing on the input image data, and outputs text data indicating the content of this image data. Has a trained model. In other words, when acquiring concept words etc. from image data, by performing image recognition using a machine learning model, it is possible to correctly recognize and convert it into text even on ordinary PCs, especially with the advancement of algorithms used in recent years. Ru. Therefore, concept words can be effectively acquired without spending a large amount of money on the hardware configuration, and accordingly, useful information can be output as idea proposal materials.

Further, the control unit 11, as an acquisition unit, extracts independent words from the input text data, and identifies and acquires concept words according to the appearance frequency of the independent words. That is, an independent word such as a noun in a text can be directly defined as a concept word. By appropriately performing processing related to natural language analysis to extract independent words and selecting key independent words based on context analysis, it becomes possible to acquire concept words easily without creating new concept words.

Further, the control unit 11 has, as an acquisition unit, a trained model that outputs concept words of text data in response to input text data. Regarding natural language analysis of text data, by using an appropriately trained machine learning model, it is possible to flexibly respond to slight flaws in the description and obtain appropriate concept words. search can be performed accurately.

Further, the control unit 11, as an acquisition unit, specifies a related term (similar word) of the acquired concept word and adds the specified related term to the search terms. As mentioned above, by using not only the concept words obtained from the input data as search terms but also related terms, such as those similar to the concept words, the breadth of the search can be expanded without significantly changing the direction of the search. This makes it possible to efficiently reduce the possibility of omissions in the search.

Further, the control unit 11, as an acquisition unit, generates a meaning vector having a direction according to the meaning of the object, and is specified by a meaning vector having a direction that satisfies a predetermined criterion with respect to the meaning vector corresponding to the concept word. Let text objects be related terms (similar words). In this way, by using semantic vectors, it is possible to objectively and quantitatively evaluate the degree of similarity and determine similar words, making it possible to stably set similar words with little variation, and search accordingly. directional deviation can be reduced.

The idea proposal support system 1 also includes a storage unit 12 that stores an auxiliary term list 122 in the server device 10. In this way, by preparing and maintaining a list of auxiliary terms in advance, it is possible to mechanically search for a large number of combinations of concept words and auxiliary terms in order as search terms.

Further, the server device 10 includes the above-mentioned control section 11 and performs each process. Therefore, the server device 10 can efficiently output data useful for idea proposal materials based on the acquired concept words, and can reduce the effort required by the user to prepare for a proposal.

In addition, the idea proposal support method of this embodiment includes an acquisition step of acquiring a concept word, a search step of performing a search using a combination of a concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term, and a search step of performing a search. and outputting data indicating characteristics corresponding to the concept word based on the result. By performing each process in this way, it becomes easier for the user to propose ideas within the company, reducing preparation time for the user and making it easier to create more persuasive materials. becomes possible.

Further, the program 121 of the present embodiment uses the computer of the server device 10 as an acquisition means for acquiring a concept word, and performs a search using a combination of a concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term. It functions as a search means and an output means for outputting data indicating characteristics corresponding to the concept word based on the search results. By installing such a program on a computer and processing it using software, it is possible to easily execute each process without preparing a special hardware configuration and output useful data for idea proposal materials. .

Note that the present invention is not limited to the embodiments described above, and various changes are possible.
For example, in the above embodiment, the concept word is set in units of words such as independent words, but in cases where it is not preferable to divide the concept words into units of words, it may be a short phrase. In this case, the search conditions may be determined to be a complete match with the short phrase.

In addition, the content of the output data does not have to be limited to graphs and charts; the output data may include part of the text data or image data obtained as a search result, as long as it indicates the content of the characteristics. .

Furthermore, in the above embodiment, it has been explained that auxiliary terms related to time and numerical values are combined, but for example, search results of multi-level evaluations categorized by alphabets, symbols, etc. can also be searched. Auxiliary terms may be set or added to the output data.

Further, the graph display is not limited to a line graph, and may be selected depending on the content (absolute value, ratio, comparison of multiple types, etc.) such as a columnar graph, bar graph, or pie graph. Furthermore, not all numerical data must be graphed or charted. It may not be graphed depending on the number of points to be displayed, or if only one image data of a graph is obtained, the image data may be output as is. Furthermore, even if a plurality of search results are obtained, the plurality of results may be output in parallel if there is a problem or difficulty in processing.

Further, in the above embodiment, the history data 212 is held and used for setting search terms, priority order, etc., but such processing does not need to be performed. Further, the history data 212 may hold only a list of search terms used in a search, and may not hold data on search results.

In addition, in the above embodiment, numerical values of multiple search results, events in time series, etc. are automatically integrated and combined. The user's approval may be required before performing the integration, or the integrated/combined data may be created so that part of the integration or combination can be easily canceled.

Further, in the above embodiment, similar words (related terms) are determined using semantic vectors, but the present invention is not limited to this. It is also possible to output only those defined as synonyms based on dictionary data or the like, or to make a determination using correlations included in the search results.

Further, in the above embodiment, the input image data and/or audio data has been described as being able to be converted into concept words, but it is not necessarily necessary to have such a function. It is also possible to always input text. In this case, it is possible to convert the image data and/or audio data into text data using other programs.

Furthermore, in the above embodiment, it has been explained that all the data that can be searched on the network is used, but the reliability can be set depending on the information source of the search results, and it can be decided not to use some of them. Alternatively, the settings may be made such that the data is used but not used for combining/integrating the data.

Furthermore, in the above embodiment, the operation reception unit 34 has been described as allowing the user to specify a search term midway through, but this may be limited to cases in which all processing is performed without including user processing. .

Furthermore, in the above embodiment, the idea proposal support system 1 has been described as having separate server devices 10, database devices 20, and terminal devices 30, but all the configurations are included and executed in a single device. Good too. Further, each device is not limited to being located within the same LAN, and may be remotely accessed from the terminal device 30 via the Internet to operate the server device 10.

Further, in the above embodiment, all processing is executed by the control unit 11 using software, but the processing of the control unit 11 may be distributed to multiple CPUs, control units of different devices, etc. . Further, part of the processing may be executed by a dedicated hardware circuit or the like.

Further, in the above embodiment, the storage unit 12 made of an HDD or a flash memory is used as an example of the computer-readable medium of the program 121 related to the idea proposal support process, but the present invention is not limited thereto. As other computer-readable media, other non-volatile memories and portable recording media such as CD-ROMs and DVD discs can be applied. Further, a carrier wave is also applied to the present invention as a medium for providing data of the program according to the present invention via a communication line.
In addition, specific details such as the configuration, processing contents, and procedures shown in the above embodiments can be changed as appropriate without departing from the spirit of the present invention.

1 Idea proposal support system 10 Server device 11 Control unit 12 Storage unit 121 Program 1211 Learned model 122 Auxiliary term list 13 Communication unit 20 Database device 21 Storage unit 211 Terminology data 212 History data 30 Terminal device 31 Control unit 32 Storage unit 33 Display Section 34 Operation reception section 35 Communication section

Claims (52)

an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output unit outputs integrated data integrated in a time series in association with the concept word.
An idea proposal support system characterized by: comprising a chart creation unit that creates chart data based on the search results related to the first type term;
The idea proposal support system according to claim 1 , wherein the output unit outputs the created chart data. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
When data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output unit outputs integrated data that is integrated with respect to the characteristic.
An idea proposal support system characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
The idea proposal support system is characterized in that the output unit combines a plurality of search results related to the second type term to generate and output combined data related to the characteristics. comprising a graph creation unit that creates a graph based on the search results related to the numerical scale,
5. The idea proposal support system according to claim 3, wherein the output unit outputs the created graph. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The idea proposal support system is characterized in that the search unit determines a priority order of the auxiliary terms to be combined with the concept word based on past search history information. 7. The idea proposal support system according to claim 6 , wherein the search unit excludes concept words included in the search history information from the search terms. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The idea proposal support system is characterized in that the search unit determines a priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result ;
a determination unit that determines the relationship between the term included in the search result and the concept word;
Equipped with
The search unit adds the term determined to be highly relevant by the determination unit to the search term and performs the search again.
An idea proposal support system characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The idea proposal support system is characterized in that the search unit performs a re-search by adding to the search condition a term that satisfies a condition regarding frequency of appearance among the terms included in the search result. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result ;
An operation reception department,
Equipped with
The search unit performs a search by adding the term accepted by the operation reception unit to the search term.
An idea proposal support system characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
Non-text data including at least one of image data and audio data and text data can be input to the acquisition unit,
The acquisition unit recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.
An idea proposal support system characterized by: The non-text data includes image data,
The idea according to claim 12 , wherein the acquisition unit includes a trained model that performs recognition processing on the input image data and outputs text data indicating the content of the image data. Proposal support system. The idea proposal support according to claim 12 or 13, wherein the acquisition unit extracts an independent word from the input text data, and specifies and acquires the concept word according to the frequency of appearance of the independent word. system. The idea proposal support system according to any one of claims 12 to 14, wherein the acquisition unit has a trained model that outputs the concept word of the text data in response to the input of the text data. . an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The acquisition unit identifies a related term to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector corresponding to the concept word. An idea proposal support system characterized in that a text object specified by a meaning vector having a direction that satisfies a predetermined criterion with respect to a meaning vector is used as the related term. The idea proposal support system according to any one of claims 3 to 15, wherein the auxiliary terms include a first type term related to time. The idea proposal according to any one of claims 1, 2, and 17, wherein the first type term includes at least a part of current events, circumstances, history, trends, eras, and chronology. support system. The idea proposal support system according to any one of claims 1, 2, and 6 to 18, wherein the auxiliary terms include second type terms related to numerical scale. 20. The idea proposal support system according to claim 3, wherein the second type term includes at least a part of share, sales, profit, and number of people. The acquisition unit specifies a related term of the acquired concept word , and adds the specified related term to the search terms. Idea proposal support system. The idea proposal support system according to any one of claims 1 to 21 , further comprising a storage unit that stores the list of auxiliary terms. an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output unit outputs integrated data integrated in a time series in association with the concept word.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
When data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output unit outputs integrated data that is integrated with respect to the characteristic.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The auxiliary term includes a second type term related to numerical scale,
The output unit combines a plurality of search results related to the second type term to generate and output combined data related to the characteristic.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit determines the priority order of the auxiliary terms to be combined with the concept word based on past search history information.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit determines the priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result ;
a determination unit that determines the relationship between the term included in the search result and the concept word;
Equipped with
The search unit adds the term determined to be highly relevant by the determination unit to the search term and performs the search again.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The search unit performs the search again by adding to the search conditions a term that satisfies a condition regarding frequency of appearance among the terms included in the search results.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result ;
An operation reception department,
Equipped with
The search unit performs a search by adding the term accepted by the operation reception unit to the search term.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
Non-text data including at least one of image data and audio data and text data can be input to the acquisition unit,
The acquisition unit recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.
An idea proposal support device characterized by: an acquisition unit that acquires concept words;
a search unit that performs a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output unit that outputs data indicating the characteristic corresponding to the concept word based on the search result;
Equipped with
The acquisition unit identifies a related term to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector corresponding to the concept word. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
An idea proposal support device characterized by: An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a type 1 term related to time,
In the output step, if data at a plurality of different timings are obtained in the search results related to the type 1 term, the integrated data integrated within the time series is output in association with the concept word.
An idea proposal support method characterized by the following. An idea proposal support method executed by a computer control unit,
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a second type term related to numerical scale,
In the output step, if data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, output integrated data that is integrated about the characteristic.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
The auxiliary term includes a second type term related to numerical scale,
In the output step, a plurality of search results related to the second type term are combined to generate and output combined data related to the characteristic.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, the priority order of the auxiliary terms to be combined with the concept word is determined based on past search history information.
An idea proposal support method characterized by the following. An idea proposal support method executed by a computer control unit,
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, a priority order of the auxiliary terms to be combined with the concept word is determined based on the relationship between the concept word and the auxiliary term.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
a determination step of determining the relationship between the term included in the search result and the concept word;
including ;
In the search step, the term determined to be highly relevant in the determination step is added to the search term and the search is performed again.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, the search is performed again by adding to the search conditions a term that satisfies the condition regarding the frequency of appearance among the terms included in the search results.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the search step, a search is performed by adding the term accepted by the operation reception unit to the search term.
An idea proposal support method characterized by the following. An idea proposal support method executed by a computer control unit,
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the acquisition step, non-text data including at least one of image data and audio data and text data can be acquired,
In the acquisition step, the content of the non-text data is recognized and converted into text data, and the concept word is acquired from the text data.
An idea proposal support method characterized by the following. An idea proposal support method executed by a control unit of a computer, the method comprising:
a retrieval step for retrieving concept words;
a search step of performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
an output step of outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
including;
In the acquisition step, a term related to the acquired concept word is identified, the identified related term is added to the search term, a meaning vector having a direction according to the meaning of the object is generated, and a meaning vector corresponding to the concept word is generated. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
An idea proposal support method characterized by the following. computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The auxiliary term includes a type 1 term related to time,
When data at a plurality of different timings are obtained in the search results related to the type 1 term, the output means outputs integrated data integrated in a time series in association with the concept word.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the search result;
function as
The auxiliary term includes a second type term related to numerical scale,
When data in a plurality of different categories related to the same characteristic is obtained in the search results related to the type 2 term, the output means outputs integrated data that is integrated with respect to the characteristic.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the search result;
function as
The auxiliary term includes a second type term related to numerical scale,
The output means combines a plurality of search results related to the second type term to generate and output combined data related to the characteristic.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the search result;
function as
The search means determines the priority order of the auxiliary terms to be combined with the concept word based on past search history information.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the search result;
function as
The search means determines the priority order of the auxiliary terms to be combined with the concept word based on the relationship between the concept word and the auxiliary term.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
determining means for determining the relevance between the terms included in the search results and the concept word;
function as
The search means adds the term determined to be highly relevant by the determination means to the search term and performs the search again.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the search result;
function as
The search means adds to the search condition a term that satisfies a condition regarding frequency of appearance among the terms included in the search result and performs the search again.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The search means performs a search by adding the term accepted by the operation reception unit to the search term.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
Non-text data including at least one of image data and audio data and text data can be input to the acquisition means,
The acquisition means recognizes the content of the non-text data, converts it into text data, and acquires the concept word from the text data.
A program characterized by: computer,
an acquisition means for acquiring concept words;
a search means for performing a search using a combination of the concept word and an auxiliary term indicating an item of characteristics of the concept word as a search term;
output means for outputting data indicating the characteristic corresponding to the concept word based on the result of the search;
function as
The acquisition means identifies a related term to the acquired concept word, adds the identified related term to the search term, generates a meaning vector having a direction according to the meaning of the object, and generates a meaning vector corresponding to the concept word. The related term is a text object specified by a meaning vector having a direction that satisfies a predetermined criterion for the meaning vector.
A program characterized by:

Images