JP6733367B2 - Task estimation device, task estimation method, and task estimation program - Google Patents

Description

The present invention relates to a technique for estimating a problem that requires verification.

Explain how to launch a business related to product development and manufacturing. One method of business development is wasteful by rapidly repeating the processes of “building a hypothesis of the product potentially needed by the customer”, “implementing the product” and “correcting the trajectory of the product and business” in the development stage. In addition, we will continue to improve products quickly while minimizing non-value elements, and lead the business to success. In this business development method, first, a person involved in the business (hereinafter referred to as a business operator) extracts a hypothetical problem related to a user such as a customer, verifies the problem, and produces a prototype product that is practically practical ( Minimum Viable Product) to be manufactured and delivered to users as quickly as possible. After that, the business operator reconstructs the problem of the original hypothesis based on the result obtained by verifying the reaction of the user to the prototype product, and corrects the trajectory of the product or business. Business owners need to spend only the minimum amount of time and money on non-essential elements (for example, creating specifications, approval documents, etc.) that do not lead to the success of the business by repeating only the minimum processes necessary to capture the demand of users. Or save waste such as labor. This makes it possible to understand the core issues without overly narrowing the scope of product development.

In order to grasp the true needs of users in such a development method, it is important to generate hypothetical tasks quickly and appropriately and obtain feedback through communication with the users. If it is determined that the hypothetical task is not appropriate, it is necessary to immediately verify the next task.

However, a group (for example, an organization that develops businesses for the elderly, etc.) having a certain common problem, a predetermined business field, or a predetermined industry (hereinafter, these are referred to as “domains” in the present application) have problems. When making a hypothesis, the business operator must comprehensively understand the information of the domain and investigate the information. In addition, the business operator needs to understand future problems of the domain and problems of fields around the domain, and it is difficult to quickly and appropriately establish a hypothesis of the problem. In addition, in the present application, the domain is also described as “information indicating the target of the problem”.

As a technique related to the above, there is disclosed a technique for expanding a target frame of problem data and creating a new idea at the time of developing a new product (see Patent Document 1). Furthermore, when designing common objects in the system design process, common objects are effectively determined from multiple domains without depending on individual domains (active areas), common object determination cycles are repeated, and common objects are updated. A technique for doing so is disclosed (see Patent Document 2).

In addition, as a related technique, there is disclosed a technique for detecting a flaw in a specification written in natural language by using morphological analysis or the like (see Patent Document 3).

Japanese Patent Laid-Open No. 2015-184971 JP-A-2004-13280 JP-A-2014-81680

In the technologies of Patent Documents 1 to 3, it is premised that the information processing device or the like appropriately receives the task data from the user or the like. However, when the task data is input from the user or the like, the task data to be input may be abstract, or the term may be inappropriate or insufficient. If the task data is not specific enough in this way, the task search results and task extraction results contain a lot of noise that is not related to the task, leading to an appropriate task, and a sufficient number of tasks. There was a problem that it could not be extracted.

The present invention has been made to solve the above problems. The main object of the present invention is to provide a task estimation device or the like that covers a wide range and extracts an appropriate task to be verified.

In order to solve the above problems, the problem estimation device according to the first aspect of the present invention is
Accepting an input of information indicating the assignment and the subject of the assignment, based on the relationship between the modifier and the modified word in the assignment, a determination means for determining whether the assignment is specific,
Among the results of searching general information with the modified word included in the problem determined to be non-specific and the information indicating the target of the problem corresponding to the problem determined to be non-specific, the modified word And an embodying means for extracting a character string including the modifier to modify the modifier as a specific task.

The problem estimation method according to the second aspect of the present invention is
Accepts input of information indicating the assignment and subject of the assignment,
Based on the relationship between modifiers and modifiers in the task, determine whether the task is specific,
Retrieval of general information using the modifiers included in the task determined to be non-specific and the information indicating the target of the task corresponding to the task determined to be non-specific as search keys,
It is provided that a character string including the modified word and the modifier that modifies the modified word is extracted as a specific problem from the search results.

The problem estimation program according to the third aspect of the present invention is
Accepts input of information indicating the assignment and subject of the assignment,
Based on the relationship between modifiers and modifiers in the task, determine whether the task is specific,
Retrieval of general information using the modifiers included in the task determined to be non-specific and the information indicating the target of the task corresponding to the task determined to be non-specific as search keys,
The computer is caused to execute, as a specific task, extracting a character string including the modified word and the modifier that modifies the modified word from the search results.

According to the present invention, it is possible to provide a task estimation device or the like that covers a wide range and extracts an appropriate task to be verified.

It is a figure which shows the structural example of the subject estimation apparatus in the 1st Embodiment of this invention. It is a figure which shows the structural example of the data of important expression. It is a figure which shows the structural example of the internal data of a domain keyword storage part. It is a flowchart which shows the operation example of the task estimation apparatus in the 1st Embodiment of this invention. It is a figure which shows the structural example of the internal data of a relationship storage part. It is a figure which shows the structural example of the internal data of a relationship storage part. It is a figure which shows the structural example of the subject estimation apparatus in the 2nd Embodiment of this invention. It is a figure which shows an example of a prioritized task list.

Next, an embodiment of the present invention will be described with reference to the drawings. In the following description of the drawings, the same or similar reference numerals are given to the same or similar parts. However, the drawings schematically show the configuration in the embodiment of the present invention. Furthermore, the embodiment of the present invention described below is an example, and can be appropriately modified within the scope of the same essence.

<First Embodiment>
(Problem estimation device)
As shown in FIG. 1, the task estimation device 100 according to the first embodiment of the present invention has a task estimation unit 1, a priority determination unit 2, an important expression storage unit 3, a general information storage unit 4, and a domain keyword storage unit. 5 and a relationship storage unit 6. The task estimation unit 1 includes a task determination unit 1a, a task materialization unit 1b, a domain extension unit 1c, and a task extension unit 1d. Note that in the present application, task estimation refers to extraction of candidates that can be tasks.

The general information storage unit 4 stores general information including a sentence (a set of character strings) used for extracting a problem to be verified, terms and character strings described in the table. Specifically, general information is a term written in a text sentence of a magazine or a newspaper, a text sentence on a website, or a chart on a website necessary for investigating a trend of a certain domain. When a diagram is described as an image, a text sentence may be extracted using an OCR (Optical character recognition) technique.

The important expression storage unit 3 stores a term (important expression term) that is likely to be included in a sentence of a task of high importance, which is included in a sentence of general information. The important expression storage unit 3 stores, for example, “important”, “maximum”, “necessary”, “required”, and the like shown in FIG. 2 as important expression terms.

The domain keyword storage unit 5 stores keywords related to a certain domain in association with each other. The domain keyword storage unit 5 includes, for example, a keyword table as shown in FIG. In the table, as the domain name of the domain, for example, "Postal", "Convenience store", "Apparel", etc., as frequently used words related to these domains, "Postcard, Postal, Yucho", "Franchise owner, ordering" ", "recycle, shop clerk", etc. are stored.

The relationship storage unit 6 is a memory for storing a table (see FIG. 5) indicating the relationship between a task and a domain when the task estimation unit 1 expands a domain or a task.

The task determination unit 1a receives input of a task and a domain, and determines whether or not the task is specific based on the relationship between the modifier and the modified word in the task. Specifically, the task determination unit 1a accepts input of text information or the like indicating “task” and “domain” from another device via an input/output interface described later. Further, the task determination unit 1a determines whether the input "task" is specific. Here, the concrete task means that the modified word (object, noun) in the task is modified by at least one modifier. For example, the task discrimination unit 1a determines whether the modified word (object, noun or noun phrase) of the input task is modified by an adjective or an adjective, and whether the adjective is modified by an adverb or an adverb phrase. To judge. The task discrimination unit 1a may discriminate a sentence into meaningful words by morphological analysis, and discriminate the part of speech and the content using a dictionary or the like. As a specific example, a case will be described in which the task determination unit 1a receives a task of “reducing the labor cost of a driver” as an input (in the following description, a specific example of a character string is in double key brackets “”). Note). In this case, since the modified word “personnel cost” is modified by the modifier “driver” (adjective phrase), the task determination unit 1a determines that the task is specific. However, when the task of “reducing personnel costs” is input, the qualified word “labor costs” is not qualified, and the task determination unit 1a determines that this task is not specific.

The task discriminating unit 1a may receive information in which the specificity of the task is determined in advance from another device, and based on this information, may determine whether or not the input task is specific.

When the task determination unit 1a determines that the input task is specific, the task determination unit 1a outputs the task to the domain extension unit 1c. When the task determination unit 1a determines that the input task is not specific, the task determination unit 1a outputs the task and the input domain to the task implementation unit 1b.

The task reification unit 1b searches the general information storage unit 4 using the modified word included in the task determined to be non-specific and the domain corresponding to the task determined to be non-specific as a search key. Furthermore, the task reification unit 1b regards a character string that includes the modified word and the modifier that modifies the modified word as a specific task among the retrieved results (sentences related to the modified word and the domain). Extract.

Specifically, the task reification unit 1b receives the task and domain determined to be unspecific from the task determination unit 1a, and extracts the modified word (noun, object) from the character string of the received task. For example, when the task materialization unit 1b receives the task “reduce labor cost”, it extracts “personnel cost” as a modifier. Next, the task materialization unit 1b searches the general information storage unit 4 using the extracted qualified word “personnel cost” and its domain (here, “physical distribution”) as a search key. At this time, instead of using all the nouns included in the task as the search key, the user may be allowed to select some nouns.

From the search results in the general information storage unit 4, the task reification unit 1b includes a search key that includes a domain (for example, “physical distribution”) and an important expression term that is stored in the important expression storage unit 3. Extract the result (sentence). At this time, since the general information storage unit 4 generally displays the search hits in descending order, the one displayed first is predicted to have a higher degree of association. Therefore, the task reification unit 1b may extract only an arbitrary number (for example, 10) of search results from the beginning.

The task reification unit 1b searches for a word that modifies the extracted qualified word (for example, “labor cost”) from any number of extracted search results, and uses the characters made up of the “qualified word” and the “modified word”. Extract columns. For example, when the modifier is “driver” and the modified word is “labor cost”, the extracted character string is “driver's cost”. The task reification unit 1b outputs a character string (for example, "personnel cost of driver") including a modifier and a modified word, which are extraction results, as a reified task. There may be a plurality of problems to be output, and the user may be allowed to select a preferred one from the plurality of problems.

The domain expansion unit 1c includes the information or the keyword indicating the target of the task stored in the keyword storage unit among the tasks which are the result of searching the general information storage unit 4 based on the search key including the input specific task. The task is extracted, and information indicating the target of the task included in the extracted task is output. Specifically, when the domain expansion unit 1c receives the embodied problem (hereinafter, referred to as a concrete problem) from the problem concrete unit 1b or the problem discrimination unit 1a, the concrete problem is expanded to another domain. Specifically, the domain expansion unit 1c searches the general information storage unit 4 using the specific task and the word "task" as search keys. For example, if the “driver's labor cost” is the task, the search key is “driver's labor cost task”.

The domain expansion unit 1c searches the search result of the general information storage unit 4 for a sentence containing the related keyword stored in the domain keyword storage unit 5, and extracts the domain name corresponding to the related keyword. The domain expansion unit 1c may search the search result of the general information storage unit 4 for a sentence including the domain name stored in the domain keyword storage unit 5. This extraction of the domain name is repeated until an arbitrary number (for example, 5) of different domain names are extracted. After that, the domain expansion unit 1c outputs the extracted domain name to the task expansion unit 1d.

The task extension unit 1d selects at least one task including a character string estimated to be a task among the results of searching the general information storage unit 4 based on the search key including the information indicating the input task target. The above is extracted. Specifically, the task extension unit 1d receives the domain name output from the domain extension unit 1c, and uses the domain name and the word "task" as a search key (for example, "taxi task") to store the general information storage unit 4 in the general information storage unit 4. Search for. The task extension unit 1d extracts a sentence including an expression estimated to be a task from the search result. The expression estimated to be a task is, for example, a sentence including a character string indicating a task such as “ga task” and “task is”. The task extension unit 1d extracts an expression estimated to be a task from the extracted sentences. For example, the task expansion unit 1d extracts a character string described before “wa task” or a character string described after “task”. The task extension unit 1d extracts an arbitrary number (for example, three) of these character strings as extended tasks. The task extension unit 1d outputs the extended task to the priority determination unit 2.

When the priority determination unit 2 receives an arbitrary number of extended tasks output by the task extension unit 1d, the priority determination unit 2 determines the priority of the extended task. For example, the priority determination unit 2 searches the general information storage unit 4 for an arbitrary number of times using the expanded task as a search key, and rearranges in the order to be verified based on the search result. The order to be verified may be, for example, the order in which the date of creation or update of the web page which is the search result is new, or the order in which the number of search hits is small. It should be noted that these are examples and the order to be verified may be realized by another method. The priority determination unit 2 outputs a task list with priorities to the outside via an input/output interface or the like described later.

(Operation of the task estimation device)
The operation of the task estimation device 100 will be described with reference to the flowchart of FIG.

In step S110, the task determination unit 1a of the task estimation apparatus 100 accepts “task” and “domain” from another device or the like via an input/output interface described later. The task determination unit 1a temporarily stores the accepted "task" and "domain" in the relationship storage unit 6 ("1. input domain", "2. input in table shown in FIG. 5"). Challenges"). In the following description, as an example, the case where the task determination unit 1a inputs the task “reduce labor cost” and the domain “physical distribution” will be described.

In step S120, the task determination unit 1a determines whether the modified word (object word, noun) of the input task is modified. In the present embodiment, as an example, the task determination unit 1a identifies whether the modified word has a modifier, that is, an adjective or an adjective, and further the adjective or adjective has an adverb or an adverb phrase. By doing so, it is determined whether or not the modified word is modified. If it is determined that the modified word is not qualified, the process proceeds to step S130. If it is determined that the modified word is modified, the process proceeds to step S140.

In step S130, the task materialization unit 1b performs processing to materialize the task “reduce labor cost” that is determined to be unspecific. Specifically, the task reification unit 1b uses the qualified word (object, noun) “labor cost” and the domain “logistics” corresponding to the qualified word as search keys (for example, “labor cost logistics”), The general information storage unit 4 is searched. From the result of the search, the task reification unit 1b extracts a sentence including the important expression term (see FIG. 2) stored in the important expression storage unit 3 and having the qualified word “personnel cost” modified. To do. The extracted sentence is, for example, “In the chart, the transportation cost per vehicle of a 4-ton vehicle per month is shown, but the labor cost of the driver accounts for 46%, which is the largest. 』And so on.

The task reification unit 1b extracts the modifier of the modified word from the extracted sentence. Furthermore, the task reification unit 1b regards a character string that combines the modified word and the extracted modifier as a reified task. For example, the task concrete part 1b shows the extracted sentence "in the chart, the transportation cost per vehicle of a 4-ton vehicle per month, but the labor cost of the driver accounts for 46%, which is the maximum. ], the "driver" that modifies the qualified word "labor cost" is extracted. Further, the task reification unit 1b regards the "personnel cost of the driver", which is a character string including the extracted modifier and modified word, as the specified problem. The task materialization unit 1b stores the materialized task in the relationship storage unit 6 (see “3. Materialized task” in the table of FIG. 5), and outputs it to the domain extension unit 1c.

In step S140, the domain extension unit 1c uses a search key (for example, “driver's labor cost task”) that combines the task “driver's labor cost” and the term “task” output by the task reification unit 1b. Then, the general information storage unit 4 is searched, and the domain is expanded using the search result. The domain expansion unit 1c includes the related keywords stored in the domain keyword storage unit 5 (see FIG. 3) in the sentence of the search result in order from the beginning of the search result (generally arranged in ascending order of the number of search hits). Investigation and output the investigation result. When the relevant keyword is included, the domain extension unit 1c determines that the domain is associated with the domain associated with the relevant keyword, and sets the domain as the extended domain. For example, the domain expansion unit 1c defines the domains "taxi", "bus", "railroad", "emergency medical care", and "garbage collection" including the related keyword "vehicle" in the domain keyword storage unit 5 as expanded domains. Output. This step is repeated until the arbitrary number of extended domains are extracted (here, 5). The domain extension unit 1c stores the extended domain in the relationship storage unit 6 (see “4. Extended domain” in the table of FIG. 5), and further outputs it to the task extension unit 1d.

In step S150, the task extension unit 1d combines the extended domains "taxi", "bus", etc. output by the domain extension unit 1c with a search key (for example, "taxi task", "bus task"). , Etc.) is used to search the general information storage unit 4. The task expansion unit 1d extracts a sentence including "task" from the search result. The number of extractions may be arbitrary.

The extracted sentences are, for example, "deterioration of the surrounding environment and congestion due to passenger waiting taxis at stations such as in front of stations or downtown" and "coexistence with welfare and nursing care taxis." ”, “Clarify what kind of bus is needed for the area, and form a bus network that allows route buses and community buses to function integrally and efficiently”, “After all, revitalize rural areas and promote tourism It seems that the issue is how to promote it. 』And so on.

The task extension unit 1d extracts a description that is considered to be a task expression from the extracted sentence. For example, the task extension unit 1d is a part of "... is a task" that corresponds to ... "deterioration of surrounding environment or traffic jam", "coexistence with welfare/nursing care taxi", and "functions integrally and efficiently. "Formation of bus network", "Measures for promoting tourism", etc. are extracted and these are expanded issues. The task expanding unit 1d stores the expanded tasks in the relationship storage unit 6 (see “5. Expanded task” in the table of FIG. 5) and further outputs the expanded tasks to the priority determining unit 2.

In step S160, the priority determination unit 2 arranges each of the expanded problems output by the problem expansion unit 1d in a new order based on the date of creation or update of the web page on which the problem was posted or the like. Change.

In step S170, the priority determination unit 2 outputs the rearranged result to another device as a priority-assigned task list (see FIG. 6) via an input/output interface described later.

As described above, the problem estimation device 100 according to the first embodiment can cover a wide range and extract an appropriate problem to be verified. This is because the task estimation unit 1 searches the general information using the modified word included in the unspecific task and the domain of the task as a search key, and the modified word that modifies the modified word and the modified word. This is because a character string including and is extracted as a specific task.

<Second Embodiment>
A task estimation device 200 according to the second embodiment of the present invention will be described with reference to FIG. 7.

The task estimation device 200 includes a determination unit 10a and a materialization unit 10b.

The determination unit 10a receives an input of a task and a domain (information indicating the target of the task), and determines whether or not the task is specific based on the relationship between the modifier and the modified word in the task.

The reification unit 10b, among the results of searching the general information using the modified word included in the problem determined to be non-specific and the domain corresponding to the problem determined to be non-specific as a search key, the modified target A character string including a word and a modifier that modifies the modified word is extracted as a specific task.

The task estimation device 200 according to the second embodiment can extract a proper task to be verified over a wide range. This is because the reification unit 10b searches the general information with the modified word included in the unspecific task and the domain of the problem as a search key, and the modified word and the modified word that modifies the modified word. This is because the character string including is extracted as a specific problem.
(Configuration of information processing device)
In each of the above-described embodiments of the present invention, each component of the task estimation device represents a block of a functional unit. Some or all of the constituent elements of the task estimation device are realized by an arbitrary combination of an information processing device 500 and a program as shown in FIG. 8, for example. The information processing device 500 includes, for example, the following configuration.

-CPU (Central Processing Unit) 501
-ROM (Read Only Memory) 502
RAM (Random Access Memory) 503
-Program 504 loaded in RAM 503
-Storage device 505 that stores the program 504
-Drive device 507 for reading and writing the recording medium 506
-Communication interface 508 connected to the communication network 509
.Input/output interface 510 for inputting/outputting data
.Bus 511 for connecting each component
Each component of the task estimation apparatus in each embodiment (each unit in the task estimation unit 1 and the like) is realized by the CPU 501 acquiring and executing the program 504 that realizes these functions. The program 504 that realizes the function of each component of the task estimation device is stored in the storage device 505 or the RAM 503 in advance, for example, and is read by the CPU 501 as necessary. The program 504 may be supplied to the CPU 501 via the communication network 509, or may be stored in the recording medium 506 in advance and the drive device 507 may read the program and supply it to the CPU 501.

There are various modifications to the method of realizing each device. For example, the task estimation device may be realized by an arbitrary combination of the information processing device 500 and the program, which are different for each component. Further, a plurality of constituent elements included in the task estimation device may be realized by an arbitrary combination of one information processing device 500 and a program.

Further, some or all of the constituent elements of the task estimation device are realized by other general-purpose or dedicated circuits, processors, etc., or a combination thereof. These may be configured by a single chip, or may be configured by a plurality of chips connected via a bus.

Some or all of the constituent elements of the task estimation device may be realized by a combination of the above-described circuits and the like and a program.

When some or all of the constituent elements of the task estimation device are realized by a plurality of information processing devices or circuits, the plurality of information processing devices or circuits may be centrally arranged or distributed. May be. For example, the information processing device, the circuit, and the like may be realized as a form in which a client and server system, a cloud computing system, and the like are connected to each other via a communication network.

1: assignment estimation unit 1a: assignment determination unit 1b: assignment implementation unit 1c: domain extension unit 1d: assignment extension unit 2: priority determination unit 3: important expression storage unit 4: general information storage unit 5: domain keyword storage unit 6: Relationship storage unit 10a: Discrimination unit 10b: Materialization unit 100: Assignment estimation device 200: Assignment estimation device 500: Information processing device 501: CPU
503: RAM
504: Program 505: Storage device 506: Recording medium 507: Drive device 508: Communication interface 509: Communication network 510: Input/output interface 511: Bus

Claims (9)

Accepting an input of information indicating the task and the subject of the task, based on the relationship between the modifier and the modified word in the task, a determination means to determine whether the task is specific,
Of the result of searching the general information using the modified word included in the problem determined to be not specific and the information indicating the target of the problem corresponding to the problem determined to be not specific as a search key, A task estimation device comprising a materialization unit that extracts a character string including the modified word and a modifier that modifies the modified word as a specific task. When the specific problem is extracted as the specific problem, from the search results, the important expression terms included in the sentence of the problem of high importance, and those that include the modifier, the extraction, the extraction Output the character string including the modified word included in the result and a modifier corresponding to the modified word as the specific problem,
The task estimation device according to claim 1. A keyword storage unit that stores at least one keyword corresponding to the information indicating the target of the task;
Among the problems that are the result of searching the general information based on the search key that includes the specific problem that is input, information that indicates the target of the problem that is stored in the keyword storage unit or a problem that includes the keyword is extracted. The task estimation device according to claim 1 or 2, further comprising a domain expansion unit that outputs information indicating a target of the task included in the extracted task. Problem expanding means for extracting at least one or more as a problem among the results of searching the general information based on the search key including the information indicating the target of the problem to be input, as a problem. The task estimation device according to any one of claims 1 to 3, further comprising: Problems estimating apparatus according to 請 Motomeko 4, further comprising a priority determining means for determining the priority of verifying the challenge acquired from the object expanding means. Accept input of information indicating the assignment and target of the assignment,
Based on the relationship between the modifier and the modifier in the task, determine whether the task is specific,
Search the general information using the modified word included in the task determined to be non-specific and the information indicating the target of the task corresponding to the task determined to be non-specific as a search key,
A task estimation method in which a computer executes, as a specific task, extracting a character string including the modified word and a modifier that modifies the modified word from the search results.
When extracting as the specific problem,
From the results of the search, to extract the important expression terms included in the sentence of the task of high importance, and those containing the modifier,
Output the character string including the modified word included in the extraction result and a modifier corresponding to the modified word as the specific problem,
The task estimation method according to claim 6. Accept input of information indicating the assignment and target of the assignment,
Based on the relationship between the modifier and the modifier in the task, determine whether the task is specific,
Search the general information using the modified word included in the task determined to be non-specific and the information indicating the target of the task corresponding to the task determined to be non-specific as a search key,
A task estimation program for causing a computer to execute, as a specific task, extracting a character string including the modified word and a modifier that modifies the modified word from the search results. When extracting as the specific problem,
From the results of the search, to extract the important expression terms included in the sentence of the task of high importance, and those containing the modifier,
Output the character string including the modified word included in the extraction result and a modifier corresponding to the modified word as the specific problem,
The task estimation program according to claim 8.

Images