JP7808166B2 - Solution proposal method, classification model generation method, and solution proposal system - Google Patents

Description

Embodiments relate to a method for proposing a solution, a method for generating a classification model, and a system for proposing a solution.

Conventionally, a set of data has been accumulated that links past problems to solutions, and when considering solutions to a new problem, attempts have been made to search the accumulated set of data for past problems that are similar to the new problem and utilize the solutions of similar past problems.
However, in some cases, the characteristics of past assignments are not fully expressed, making it impossible to find appropriate past assignments similar to the new assignment from the accumulated set data.

Japanese Patent Application Laid-Open No. 2000-276487

The purpose of the embodiment is to provide a solution proposal method and proposal system that can detect appropriate past problems that are similar to a new problem, and a method for generating a classification model to be used in such a solution proposal method.

A method for proposing a solution according to an embodiment includes the steps of: acquiring a new assignment; using a classification model trained with training data in which assignment tags corresponding to multiple past assignments are classified into multiple categories according to their mutual relevance, estimating which of the multiple categories the new assignment will be classified into; searching for a past assignment similar to the new assignment among the multiple past assignments within the estimated category; and outputting a solution associated with the searched similar past assignment; each of the past assignments is entered into one of the assignment items in at least one hierarchical data set in which assignment items and means items are arranged in a hierarchical structure; and each of the assignment tags is determined based on information entered into each of the past assignments and into items in the hierarchical data that have a parent-child relationship with the assignment item into which the each of the past assignments is entered.

FIG. 1 is a block diagram showing a system for proposing a solution according to a first embodiment and a system for generating a classification model used in the system. FIG. 10 is a diagram illustrating an example of hierarchical data. 1 is a flowchart illustrating a method for generating a classification model according to the first embodiment. FIG. 1 illustrates a method for determining issue tags. FIG. 1 illustrates a method for determining issue tags. FIG. 10 illustrates a method for classifying multiple assignment tags into multiple categories. 1 is a flowchart showing a method for proposing a solution according to the first embodiment. FIG. 1 illustrates a method for estimating the category into which a new task falls. 10 is a graph showing the maximum value of the relevance degree for each category. FIG. 10 is a diagram showing a method for searching for similar past assignments. FIG. 10 is a diagram illustrating an example of an output. Figure 12(a) is a diagram showing a method for estimating a category into which a new assignment will be classified in the system for proposing a solution relating to the second embodiment, and Figure 12(b) is a diagram showing a method for extracting a past assignment that is similar to a new assignment from among multiple past assignments in the system for proposing a solution relating to the second embodiment. FIG. 11 is a diagram showing a search condition setting screen of the solution proposing system according to the second embodiment. FIG. 10 is a diagram illustrating an output of a solution proposing system according to the second embodiment. FIG. 11 is a diagram showing a method for determining a problem tag in the solution proposal system according to the third embodiment. FIG. 13 is a diagram showing past data according to the fourth embodiment.

First Embodiment
First, the first embodiment will be described.
(Outline of the proposed solution system)
FIG. 1 is a block diagram showing a system for proposing a solution according to this embodiment and a system for generating a classification model used in the system.
A proposed system 1 (hereinafter simply referred to as "proposed system 1") of a solution according to this embodiment is used, for example, in an organization consisting of multiple domains. A "domain" is a business area in which an organization such as a company develops its business.

When a new issue i1 is input from a terminal in one of the domains, the proposed system 1 searches for a past issue a11 similar to the new issue i1 from among multiple past issues stored within the organization, and outputs a solution a12 for the similar past issue a11. This allows past knowledge to be utilized regardless of the domain when considering a solution to the new issue i1 in each domain. Note that the proposed system 1 may also be used within a single domain.

The proposed system 1 is based on at least one piece of hierarchical data D1 (past data) provided from each domain, a classification model D5 generated by a generation system 2 using the hierarchical data D1, and set data D2. Before describing the configuration of the proposed system 1 in detail, we will first provide a detailed description of the hierarchical data D1 and the generation system 2.

(hierarchical data)
FIG. 2 is a diagram illustrating hierarchical data.
The hierarchical data D1 is data that breaks down knowledge that was considered in the past when realizing an object such as technology or a product into problems that arose when realizing the object and means that can solve those problems, and links the means to the problems. By creating the hierarchical data D1, it is possible to organize the past knowledge that was considered when realizing the object.

In hierarchical data D1, task item a1 and means item a2 are arranged in a hierarchical structure. A "hierarchical structure" refers to a structure having multiple levels arranged from top to bottom, with item Y, which is a prerequisite for a certain item X, placed in the level immediately below the level in which item X is placed, and the lower-level item Y is linked to the higher-level item X. Hereinafter, this relationship between item X and item Y is referred to as a "parent-child relationship." Furthermore, item X is referred to as a "parent item" from the perspective of item Y, and item Y is referred to as a "child item" from the perspective of item X. Furthermore, if item Z is placed in the level immediately below the level in which item Y is placed and is linked to item Y, item Z is referred to as a "grandchild item" from the perspective of item X. Hierarchical data D1 may also be configured as a hierarchical structure that includes items other than task item a1 and means item a2. In such a case, multiple items, including task item a1 and means item a2, are arranged in the hierarchical data D1 to form a hierarchical structure.

In the example of FIG. 2, target item a0 is placed at the top level (root) of hierarchical data D1. For example, target technology, products, etc. are input into target item a0. A child item of target item a0 is problem item a1 that arises when realizing target item a0. A child item of problem item a1 is means item a2 that can solve problem item a1, or another problem item a1 obtained by subdividing problem item a1 into multiple subordinate concepts. A child item of means item a2 is problem item a1 that arises when means item a2 is adopted, or another means item a2 obtained by subdividing means item a2 into multiple subordinate concepts.
In the hierarchical data D1, when a plurality of child items are linked to one parent item, the child items are of the same type.

The proposal system 1 and the generation system 2 treat an assignment that is input to any assignment item a1 of at least one piece of hierarchical data D1 as a "past assignment a11."
Furthermore, the proposal system 1 and the generation system 2 treat, for example, the means that were actually adopted to solve the past problem a11 among the means input into the means item a2 of the descendants such as children and grandchildren of the problem item a1 in which the past problem a11 was input as the "solution means a12." Note that the proposal system 1 and the generation system 2 may also treat, for example, all the means, including the means that were not adopted, among the means input into the means item a2 of the descendants of the problem item a1 in which the past problem a11 was input as the "solution means a12."

(Configuration of a classification model generation system)
Next, the configuration of the generation system 2 will be described.
As shown in FIG. 1, the generation system 2 includes a tag determination unit 21, a data creation unit 22, and a model generation unit 23.

The tag determination unit 21 determines the assignment tag c4 corresponding to each past assignment a11 based on the information entered for each past assignment a11 and for items in the hierarchical data D1 that have a parent-child relationship with the assignment item a1 into which the past assignment a11 was entered. The "items in a parent-child relationship" are the parent item b1 and child item b2 of the assignment item a1 into which the past assignment a11 was entered, as shown in Figure 2.

The assignment tag c4 is determined based on, for example, words extracted from the text of each past assignment a11 and the text entered in the parent item b1 and child item b2, as well as synonyms of the extracted words. As shown in FIG. 1, the tag determination unit 21 uses, for example, ontology D3 to search for synonyms and determine the assignment tag c4.

Ontology D3 systematizes the relationships between concepts used in multiple domains. Even if a problem has a similar meaning, it may be expressed using different words in different domains. Ontology D3 is a concept dictionary that incorporates, for example, the relationships between words used in each domain, and systematizes the hierarchical and synonymous relationships between words.

The data creation unit 22 creates training data D4 in which multiple task tags c4 are classified into multiple categories according to their mutual relevance. Specifically, the data creation unit 22 uses ontology D3 to divide the multiple task tags c4 into task tag groups that can be grouped under a unified concept across domains. The data creation unit 22 treats one task tag group as one category. In other words, each category contains multiple task tags c4 that can be grouped under a unified concept across domains. Therefore, a "domain" is an organization's business area, such as "train air conditioning business" or "nuclear power business," while a "category" is a task performed in common across multiple domains, such as "procurement" or "installation."

The data creation unit 22 also creates set data D2, which is organized by category, by linking past issues a11 and solutions a12 extracted from the hierarchical data D1 to issue tags c4. The set data D2 is provided to the search unit 13 and output unit 14 of the proposal system 1, which will be described later.

The model generation unit 23 generates a classification model D5 using the training data D4 created by the data creation unit 22. The generated classification model D5 is provided to the estimation unit 12 of the proposal system 1, which will be described later, and is used by the estimation unit 12 to estimate the category into which the new task i1 will be classified.

Classification model D5 is a text classification model. Classification model D5 is, for example, a multi-class classification model that allows one new task i1 to be classified into multiple categories. The method used for classification model D5 is, for example, a neural network. An example of a text classification model using a neural network is fastText, an open source software from Facebook, Inc.

However, the method used for classification model D5 is not limited to the above, and may be, for example, a support vector machine (SVM), k-nearest neighbor method, logistic regression, etc. The classification model may also be an exclusive classification model that classifies one new task i1 into only one category.

The classification model D5 created in this way can, for example, take any phrase as input and output the relevance of the input phrase in each category.

(Configuration of the proposed solution system)
Next, the configuration of the proposed system 1 will be described.
The proposed system 1 includes an input unit 11, an estimation unit 12, a search unit 13, and an output unit 14.

The input unit 11 accepts input to the proposal system 1. Users in each domain use their terminals to input new tasks i1, set search conditions, and issue search instructions to the proposal system 1. The input unit 11 receives search instructions from the terminal and acquires new tasks i1, search conditions, etc.

The estimation unit 12 uses the classification model D5 provided by the generation system 2 to estimate the category into which the new task i1 is classified from among multiple categories.

The search unit 13 references set data D2, in which multiple past assignments a11 are organized by category, and searches for past assignments a11 that are similar to the new assignment i1 among the multiple past assignments a11 within the estimated category. In this way, past assignments a11 similar to the new assignment i1 are searched for within categories to which highly related past assignments a11 belong. This makes it possible to detect appropriate past assignments a11 that are similar to the new assignment i1.

When searching for past assignments a11 similar to the new assignment i1, the search unit 13 may use a vectorization tool D6 that vectorizes text. Examples of vectorization tools D6 include Word2Vec and Doc2Vec. The vectorization tool D6 has already trained using, for example, the text entered in each item in the hierarchical data D1 as training data.

The search unit 13 uses the vectorization tool D6 to vectorize the text of the new assignment i1 and the text of each past assignment a11 belonging to the estimated category. For example, for each category, the search unit 13 determines that the past assignment a11 represented by a vector that is closest to the vector representing the new assignment i1 is the past assignment a11 similar to the new assignment i1. Note that the search unit 13 may also detect past assignments a11 represented by vectors whose distance from the vector representing the new assignment i1 is equal to or less than a threshold, and determine the detected past assignments a11 to be the past assignments a11 similar to the new assignment i1.

The output unit 14 references the set data D2 provided by the generation system 2 and outputs a solution a12 for a past assignment a11 that is similar to the new assignment i1.

Proposal system 1 and generation system 2 are realized, for example, by a server having a CPU (Central Processing Unit) and memory and connected to terminals in their respective domains. The server may be composed of a single computer, or multiple computers. Proposal system 1 and generation system 2 may also be realized by separate computers. Furthermore, the hardware that realizes the proposal system and generation system is not limited to a server. For example, the proposal system and generation system may be realized by a computer that performs processing using various data, such as new tasks and past data, uploaded to the cloud. Furthermore, the hardware that realizes the proposal system and the hardware that realizes the generation system may be different.

(Method for generating a classification model)
FIG. 3 is a flowchart showing a method for generating a classification model according to this embodiment.
4 and 5 are diagrams illustrating a method for determining issue tags.
Next, a method for generating a classification model according to this embodiment will be described using a specific example.
First, as shown in Figures 3 to 5, the tag determination unit 21 of the generation system 2 determines an assignment tag c4 corresponding to each past assignment a11 based on the information entered into each past assignment a11 and into items b1 and b2 that have a parent-child relationship with the assignment item a1 into which each past assignment a11 was entered within the hierarchical data D1 (step S11).

More specifically, as shown in FIG. 4, the tag determination unit 21 performs morphological analysis on the sentence in past assignment a11, the sentence entered in parent item b1, and the sentence entered in child item b2, and extracts words (independent words) from each sentence, excluding particles and auxiliary verbs (step S11a). Hereinafter, the extracted words will be referred to as "extracted words c1." To explain using the example of FIG. 4, the sentence in past assignment a11 is "product durability." The sentence entered in parent item b1 is "procurement work for semiconductor application products." The sentences entered in the five child items b2 are "appropriate design," "appropriate material selection," "appropriate manufacturing," "appropriate inspection," and "appropriate shipping and transportation." The independent words in these sentences are extracted words c1.

Next, the tag determination unit 21 references the ontology D3 and adds synonyms c2 for each extracted word c1 (step S11b).

Next, as shown in FIG. 5, the tag determination unit 21 determines a representative word c3 for the plurality of extracted words c1 and the plurality of synonyms c2 using the ontology D3 (step S11c).
Specifically, first, the tag determination unit 21 divides the extracted words c1 and the synonyms c2 into word groups g that can be grouped under a unified concept using the ontology D3. Next, the tag determination unit 21 selects several word groups g from the word groups g. The selection method is not particularly limited, but for example, word groups g with low priority from the perspective of expressing the problem or the like may be removed, and then the top word groups g containing the largest number of words may be selected.

Next, the tag determination unit 21 determines a representative word c3 for each selected word group g by aggregating the extracted words c1 and synonyms c2 contained therein. Each representative word c3 may be, for example, a hypernym of the extracted words c1 and synonyms c2 contained in each word group g, or any of the extracted words c1 and synonyms c2 contained in each word group g. Note that in the example of Figure 5, there are six representative words c3, but the number of representative words c3 is not particularly limited.

Next, the tag determination unit 21 determines a task tag c4 based on the determined representative words c3 (step S11d). The task tag c4 may be a combination of all the representative words c3, or it may be a word or combination of words that aggregates the determined representative words c3. For convenience, in the figure, each task tag c4 is represented by a string of characters consisting of a "Y" followed by a number.

The tag determination unit 21 determines the task tag c4 based on the information entered not only in past task a11 but also in its parent-child items b1 and b2. For example, using Figures 4 and 5 as an example, task tag c4 includes features such as "procurement," "design," "manufacturing," and "transportation" that are not expressed in the sentence "product durability" in past task a11. In this way, task tag c4 includes features of information from higher and lower hierarchical levels of past task a11.

The tag determination unit 21 also determines assignment tags c4 based on extracted words c1 and synonyms c2 of extracted words c1 extracted from the sentences of each past assignment a11 and the sentences entered in the parent-child items b1 and b2. This allows an appropriate assignment tag c4 to be assigned to each past assignment a11, without being limited to the expressions in the sentences of the past assignment a11 and the sentences entered in the parent-child items b1 and b2.

Note that the method for determining the task tag c4 is not limited to the above. For example, the task tag c4 may not be determined mechanically by the tag determination unit 21 using ontology D3 or the like, but may be determined manually by an expert, or may be determined using a method that combines mechanical and manual methods.

FIG. 6 is a diagram showing a method for classifying multiple issue tags into multiple categories.
As shown in FIG. 6, in the above-described step S11, an assignment tag c4 is assigned to each past assignment a11 in each domain.

Next, the data creation unit 22 creates teacher data D4 in which a plurality of task tags c4 are classified into a plurality of categories according to their mutual relevance (step S12).
In step S12, for example, the data creation unit 22 uses the ontology D3 to divide multiple issue tags c4 into issue tag groups h that can be organized under a unified concept across multiple domains. The data creation unit 22 treats one issue tag group h as one category. Note that in this case, for example, one issue tag c4 may belong to multiple categories. In the example shown in Figure 6, the categories include "procurement system,""installationsystem,""productsystem,""maintenancesystem,""designsystem,""productionsystem," and "transport system."

Note that the method for creating the training data D4 is not limited to the above. For example, the task tags c4 may not be mechanically categorized by the data creation unit 22 using ontology D3 or the like, but may instead be manually categorized by an expert or the like, or may be categorized using a method that combines mechanical and manual methods.

The data creation unit 22 also creates set data D2 (see FIG. 10) in which the assignment tag c4 is linked to the past assignment a11 and the solution a12 and organized by category. The data creation unit 22 provides the set data D2 to the search unit 13.

Next, the model generation unit 23 generates a classification model D5 using the training data D4 (step S13). The classification model D5 generated by the model generation unit 23 is provided to the estimation unit 12 of the proposed system 1.

When generating the classification model D5 using the training data D4, all of the training data D4 created by the data creation unit 22 may be used to train the classification model D5, or only a portion of the training data D4 created by the data creation unit 22 may be used to train the classification model D5.

(Method of proposing a solution)
FIG. 7 is a flowchart showing a method for proposing a solution according to this embodiment.
Next, a method for proposing a solution according to this embodiment will be described.
First, a user uses a terminal to input a new assignment i1, set search conditions, and issue a command to start a search to the proposal system 1. The input unit 11 receives the command to start a search and acquires the new assignment i1, the search conditions, etc. (step S21).

Next, the estimation unit 12 uses the classification model D5 trained in step S13 to estimate the category into which the new task i1 is classified from among multiple categories (step S22).

FIG. 8 is a diagram illustrating a method for estimating a category into which a new task falls.
For example, the estimation unit 12 extracts a plurality of partial words i2 by shifting their positions from the sentence of the new task i1. For example, the estimation unit 12 sequentially extracts partial words i2 each consisting of three characters while shifting the extraction start position by one character at a time. However, the number of characters in the partial word i2 and the number of characters to be shifted are not limited to those described above.

Next, the estimation unit 12 inputs the extracted multiple partial words i2 into the classification model D5, and causes the classification model D5 to output the relevance i3 between each partial word i2 and each category. For example, the classification model D5 calculates a vertical vector V for each partial word i2, with the relevance i3 with each category as its component. Hereinafter, the horizontal arrangement of the vertical vectors V for each partial word i2 will be referred to as the relevance matrix M.

Next, the estimation unit 12 calculates the maximum value of the relevance i3 for each category in the relevance matrix M, and determines whether the maximum value calculated for each category is equal to or greater than a threshold. In the example of Figure 8, the maximum value of the relevance i3 for the "procurement" category is 0.77. The maximum value of the relevance i3 for the "installation" category is 0.99. The maximum value of the relevance i3 for the "product" category is 0.11. The maximum value of the relevance i3 for the "maintenance" category is 0.73.

FIG. 9 is a graph showing the maximum relevance values for each category.
As shown in Figure 9, the estimation unit 12 estimates that the category in which the new issue i1 will be classified is the category in which the maximum value of the relevance degree i3 is equal to or greater than the threshold. In the example of Figure 9, the maximum values of the relevance degree i3 of "installation system", "maintenance system", and "procurement system" exceed the threshold. Therefore, the estimation unit 12 estimates that the new issue i1 will be classified into "installation system", "maintenance system", and "procurement system". Note that the threshold may be set to an arbitrary value by the user in step S21, for example.

In this way, multiple partial words i2 are extracted from the sentence of new task i1 at shifted positions. Then, classification model D5 takes the extracted multiple partial words i2 as input and outputs the relevance i3 between each partial word i2 and each category. Then, based on the relevance i3, the category into which new task i1 will be classified is estimated. At least one partial word i2 contains part of a group of words located nearby in the sentence. As a result, it is possible to estimate a contextually related category for the sentence of new task i1. Note that the method of estimating the category into which new task i1 will be classified is not limited to the above.

FIG. 10 is a diagram showing a method for searching for similar past assignments.
Next, as shown in FIG. 10, the search unit 13 refers to the set data D2 and searches for past assignments a11 similar to the new assignment i1 among the multiple past assignments a11 in each estimated category (step S23).

In step S23, the search unit 13 vectorizes the text of the new assignment i1 and the text of the past assignment a11 within each estimated category, for example, using the vectorization tool D6.

Next, the search unit 13 detects, for example, within each category, the past assignment a11 represented by the vector closest to the vector representing the new assignment i1, and determines that the detected past assignment a11 is the past assignment a11 that is similar to the new assignment i1.

In the example shown in FIG. 10, a past issue a11 with an issue tag c4 of "Y22" in the "Installation" category is detected as a past issue a11 similar to the new issue i1. A past issue a11 with an issue tag c4 of "Y21" in the "Maintenance" category is detected as a past issue a11 similar to the new issue i1. A past issue a11 with an issue tag c4 of "Y33" in the "Procurement" category is detected as a past issue a11 similar to the new issue i1. For example, the search unit 13 may search for a past issue a11 similar to the new issue i1 from among multiple past issues a11 within a category based on the issue tag c4. In this case, too, the vectorization tool D6 may be used to vectorize the issue tag c4 corresponding to the new issue i1 and the issue tag c4 corresponding to the estimated past issues a11 in each category.

In this way, past assignments a11 similar to the new assignment i1 are searched for within a category to which highly related past assignments a11 belong. This makes it possible to detect appropriate past assignments a11 similar to the new assignment i1. As mentioned above, the assignment tag c4 contains information about the higher and lower hierarchical levels of the past assignment a11. Therefore, each past assignment a11 is classified into an appropriate category using the assignment tag c4. Note that the method of searching for past assignments a11 similar to the new assignment i1 is not limited to the above.

FIG. 11 is a diagram illustrating an example of the output.
Next, the output unit 14 refers to the set data D2 and outputs the solutions a12 of the similar past assignments a11 for each estimated category (step S24).
The output unit 14 outputs, for example, the estimated categories, similar past assignments a11 in each category, and their solutions a12. There may be one or more similar past assignments a11 per category. The user can consider solutions to the new assignment i1 by referring to the output solutions a12.

The classification model D5 only needs to be generated at least once before the proposal system 1 is used for the first time. For example, the classification model D5 may be generated only once, and the proposal system 1 may use the same classification model D5 each time it estimates the category into which a new task i1 will be classified. Furthermore, the classification model D5 may be regenerated, for example, when new hierarchical data D1 is provided or when the contents of existing hierarchical data D1 are updated.

Next, the effects of this embodiment will be described.
In this embodiment, a past assignment a11 similar to the new assignment i1 is searched for within the category into which the new assignment i1 is estimated to be classified. Therefore, a past assignment a11 similar to the new assignment i1 is searched for within a category to which highly related past assignments a11 belong. As a result, an appropriate past assignment a11 similar to the new assignment i1 can be detected.

In addition, in this embodiment, each past assignment a11 is entered into one of the assignment items a1 in at least one hierarchical data D1, in which assignment items a1 and means items a2 are arranged in a hierarchical structure. The assignment tag c4 is determined based on the information entered into each past assignment a11 and into the items b1 and b2 that have a parent-child relationship with the assignment item a1 to which the past assignment a11 is entered in the hierarchical data D1. Therefore, the assignment tag c4 contains characteristics of the information from the higher and lower hierarchical levels of the past assignment a11. Therefore, each past assignment a11 can be classified into an appropriate category using the assignment tag c4.

Furthermore, in this embodiment, each assignment tag c4 is determined based on extracted words c1 and synonyms c2 of extracted words c1 extracted from the sentences of each past assignment a11 and the sentences entered in items b1 and b2 in a parent-child relationship. This allows an appropriate assignment tag c4 to be assigned to each past assignment a11, without being limited to the expressions in the sentences of the past assignment a11.

In addition, in this embodiment, multiple partial words i2 are extracted from the sentence of new task i1 at shifted positions. Then, classification model D5 receives the extracted multiple partial words i2 as input and outputs the relevance i3 between each partial word i2 and each category. Then, based on the output relevance i3, the category into which new task i1 will be classified is estimated from among the multiple categories. At least one of the partial words i2 contains part of a group of words located in a nearby position in the sentence of new task i1. As a result, it is possible to estimate contextually related categories for the sentence of new task i1.

Furthermore, in this embodiment, each category contains multiple assignment tags c4 that can be grouped under a unified concept across multiple domains. As a result, it is possible to search for appropriate past assignments a11 that are similar to new assignments i1, regardless of the domain.

The above-described embodiment makes it possible to realize a method and system for proposing a solution means a12 that can detect an appropriate past task a11 that is similar to a new task i1, as well as a method for generating a classification model D5 that can be used in the method for proposing such a solution means a12.

Second Embodiment
Next, a second embodiment will be described.
Figure 12(a) is a diagram showing a method for estimating the category into which a new assignment will be classified in the system for proposing a solution according to this embodiment, and Figure 12(b) is a diagram showing a method for extracting past assignments similar to a new assignment from among multiple past assignments in the system for proposing a solution according to this embodiment.
FIG. 13 is a diagram showing a search condition setting screen of the solution proposing system according to this embodiment.
FIG. 14 is a diagram showing the output of the solution proposal system according to this embodiment.
In the following description, only the differences from the first embodiment will be described in principle, and the matters other than those described below are the same as those in the first embodiment.

As shown in FIG. 12(a), the estimation unit 12 is capable of calculating the degree of association between new assignment i1 and multiple categories, as in the first embodiment. The estimation unit 12 then estimates, from among the multiple categories, a category whose degree of association is equal to or greater than a first threshold TH1 as the category into which new assignment i1 will be classified.

Furthermore, as shown in FIG. 12(b), the search unit 13 can calculate the similarity between the new assignment i1 and a similar past assignment a11. Then, the search unit 13 extracts, as similar past assignments a11, past assignments a11 that are classified into the estimated category from among the multiple past assignments a11 and whose similarity to the new assignment i1 is equal to or greater than a second threshold TH2. The similarity is not particularly limited as long as it represents the degree of similarity between the new assignment i1 and the past assignment a11, but examples include cosine similarity or the distance between the new assignment i1 and the past assignment a11 in semantic space.

As shown in FIG. 13, the input unit 11 also accepts the user's settings of the first threshold TH1 and second threshold TH2 when inputting new assignment i1. For example, before a search, a screen such as the one shown in FIG. 13 is displayed on the computer operated by the user. The user inputs new assignment i1 via an operation unit such as the computer's keyboard or mouse, sets the first threshold TH1 and second threshold TH2, and presses the search button 210. This starts a search using the set conditions.

As shown in FIG. 14, when outputting the solution a12 of the similar past assignment a11, the output unit 14 also outputs the relevance k1 between the estimated category and the new assignment i1. Furthermore, the output unit 14 outputs, as reference information, the multiple categories with the highest relevance levels with respect to the new assignment i1 that are less than the first threshold TH1, along with their relevance levels k2. This allows the user to check the relevance levels k2 of categories into which the new assignment i1 was not estimated to be classified, and determine whether the first threshold TH1 is appropriate. Note that, as reference information, it is also possible to output only the category with the highest relevance level and its relevance level among categories with relevance levels less than the first threshold.

When outputting the solution a12 for the similar past assignment a11, the output unit 14 also outputs the similarity k3 between the new assignment i1 and the similar past assignment a11. Furthermore, the output unit 14 outputs, as reference information, the past assignment a11 with the highest similarity k3 to the new assignment i1 whose similarity k3 is less than the second threshold TH2, and its similarity k4. This allows the user to check the similarity k4 of the past assignment a11 that was not extracted, and determine whether the second threshold TH2 is appropriate. Note that, among the past assignments with similarity less than the second threshold, only the past assignment with the highest similarity and its relevance may be output as reference information.

The input unit 11 is configured to be able to accept changes to the first threshold TH1 and the second threshold TH2 even after outputting the solution a12 for a similar past problem a11. For example, when a screen like the one shown in FIG. 14 is displayed on a computer operated by a user, the user can reset the first threshold TH1 and the second threshold TH2 via an operation unit such as the computer's keyboard or mouse, and press the search again button 220. This will start a search again using the set conditions.

As described above, the output unit 14 outputs the relevance k1 between the estimated category and new task i1. This allows the user to understand the relevance k1 between the estimated category and new task i1.

The output unit 14 also outputs the relevance k2 of the category with the highest relevance among the multiple categories whose relevance to the new assignment i1 is less than the first threshold TH1. The input unit 11 is also configured to accept changes to the first threshold TH1. This allows the user to determine whether the first threshold TH1 is an appropriate value, and, if necessary, change the value of the first threshold TH1 to increase or decrease the number of categories into which the new assignment i1 is estimated to be classified.

The search unit 13 also outputs the similarity k3 between the extracted similar past assignment a11 and the new assignment i1. This allows the user to understand the similarity k3 between the extracted past assignment a11 and the new assignment i1.

The output unit 14 also outputs the similarity k4 of the past assignment a11 with the greatest similarity among the multiple past assignments a11 whose similarity to the new assignment i1 is less than the second threshold TH2. The input unit 11 is also capable of accepting changes to the second threshold TH2. Therefore, the user can determine whether the second threshold TH2 is an appropriate value, and change the value of the second threshold TH2 as necessary to increase or decrease the number of similar past assignments a11 that are extracted.

Third Embodiment
Next, a third embodiment will be described.
FIG. 15 is a diagram showing a method for determining a problem tag in the solution proposal system according to this embodiment.
In the first embodiment, the assignment tag c4 of the past assignment a11 is determined based on the past assignment a11 and the information entered in the items in the hierarchical data D1 that have a parent-child relationship with the past assignment a11. In contrast, in the third embodiment, the assignment tag c4 of the past assignment a11 is determined based on the past assignment a11 and the information related to the past assignment a11 in the domain 300 to which the past assignment a11 belongs in the hierarchical data D1.

As in the first embodiment, multiple hierarchical data D1 are provided from multiple domains 300. Therefore, in the multiple hierarchical data D1, each past assignment a11 belongs to one of the domains 300. As shown in FIG. 15, when determining the assignment tag c4 for past assignment a11, "Check the quality of incoming goods," which belongs to a certain domain 300, the tag determination unit 21 first extracts information related to past assignment a11 within the same domain 300 from the multiple hierarchical data D1. The extracted information is not particularly limited, but may include, for example, a sentence entered in assignment item a1, a sentence entered in means item a2, or both sentences entered in assignment item a1 and means item a2. In FIG. 15, for example, the sentence entered in assignment item a1, "Reduce variation by changing the production process or recipe," and the sentence entered in assignment item a1, "Introduce sampling inspections for redundant inspections and reliability," are extracted as related information.

The method for extracting information related to past assignments a11 is not particularly limited, but examples include mechanical methods using text vectorization tools D6 such as Word2Vec, manual methods such as experts extracting related information, or a combination of mechanical and manual methods. When using the vectorization tool D6, information that belongs to the same domain 300 from multiple hierarchical data D1 and is close in semantic space is extracted.

The subsequent steps are the same as in the first embodiment, where words are extracted from the sentences "check the quality of incoming products," "reduce variability by changing the production process or recipe," and "implement sampling inspections for redundant inspections and reliability," and synonyms of the extracted words are added to determine the final issue tag c4.

As explained above, the information used to determine the assignment tag c4 of past assignment a11 in hierarchical data D1 is not particularly limited to the information entered in an item that has a parent-child relationship with the assignment item a1 to which past assignment a11 is entered, but is not particularly limited as long as it belongs to the same domain to which past assignment a11 belongs in hierarchical data D1 and is information related to past assignment a11.

<Fourth embodiment>
Next, a fourth embodiment will be described.
FIG. 16 is a diagram showing past data according to this embodiment.
In the first, second, and third embodiments, a past assignment a11 is input, and the past data used to determine the assignment tag c4 is hierarchical data D1. In contrast, in this embodiment, a past assignment a11 is input, and the past data used to determine the assignment tag c4 is table data D41.

Table data D41 is tabular data in which multiple solutions a12 are linked to multiple past issues a11. In this case, the issue tag c4 for past issue a11 is determined based on the past issue a11 and information in table data D41 that belongs to the same domain 300 as past issue a11 and is related to past issue a11, as in the third embodiment. For example, when determining the issue tag c4 for past issue a11 "Check the quality of incoming products" that belongs to a certain domain 300, the past issue a11 "Reducing variation by changing the production process or recipe" and the past issue a11 "Implementing sampling inspections for redundant inspections and reliability" are extracted as related information, as in the third embodiment. However, as in the third embodiment, the solution a12 may also be extracted as related information.

As explained above, past assignments a11 are input, and the past data used to determine assignment tags c4 is not limited to hierarchical data D1, but may be data in other formats such as table data D41, as long as the past assignments a11 are linked to solution methods a12.

Although the embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments may be embodied in a variety of other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. These embodiments and their variations are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as set forth in the claims.

1: Proposal system 2: Generation system 11: Input unit 12: Estimation unit 13: Search unit 14: Output unit 21: Tag determination unit 22: Data creation unit 23: Model generation unit D1: Hierarchical data (past data)
D2: Set data D3: Ontology D4: Training data D5: Classification model D6: Vectorization tool D41: Table data M: Relevance matrix V: Vertical vector a0: Target item a1: Problem item a11: Past problem a2: Means item a12: Solution means b1: Parent item b2: Child item c1: Extracted term c2: Synonym c3: Representative term c4: Problem tag g: Word group h: Problem tag group i1: New problem i2: Partial word i3, k1, k2: Relevance k3, k4: Similarity TH1: First threshold TH2: Second threshold

Claims (11)

It runs on a computer
acquiring new assignments;
using a classification model to estimate a category among a plurality of categories into which the new task will be classified;
A step of searching for a past assignment that is classified into the plurality of categories and is similar to the new assignment and that is classified into the estimated category from among the plurality of past assignments;
outputting solutions associated with the searched similar past problems;
Equipped with
In the step of estimating a category into which the new assignment will be classified, a category among the plurality of categories whose relevance to the new assignment is equal to or greater than a first threshold is estimated as the category into which the new assignment will be classified;
In the step of outputting the solution, a degree of association between the estimated category and the new problem is output;
In the step of outputting the solution, a degree of relevance of a category, of the plurality of categories, whose degree of relevance with the new problem is less than the first threshold is output;
A method proposed by the solution means , in which the first threshold value can be changed by a user . In the step of searching for similar past assignments, past assignments that are classified into the estimated category from among the plurality of past assignments and have a similarity to the new assignment that is equal to or greater than a second threshold are extracted as the similar past assignments;
2. The method for proposing a solution according to claim 1 , wherein the step of outputting the solution comprises outputting a similarity between the extracted similar past assignment and the new assignment. It runs on a computer
acquiring new assignments;
using a classification model to estimate a category among a plurality of categories into which the new task will be classified;
A step of searching for a past assignment that is classified into the plurality of categories and is similar to the new assignment and that is classified into the estimated category from among the plurality of past assignments;
outputting solutions associated with the searched similar past problems;
Equipped with
In the step of searching for similar past assignments, past assignments that are classified into the estimated category from among the plurality of past assignments and have a similarity to the new assignment that is equal to or greater than a second threshold are extracted as the similar past assignments;
In the step of outputting the solution, a similarity between the extracted similar past problem and the new problem is output;
In the step of outputting the solution, a similarity of a past assignment that is classified into the estimated category and has a similarity to the new assignment that is less than the second threshold is output from among the plurality of past assignments;
The second threshold value can be changed by the user . It runs on a computer
acquiring new assignments;
using a classification model to estimate a category among a plurality of categories into which the new task will be classified;
A step of searching for a past assignment that is classified into the plurality of categories and is similar to the new assignment and that is classified into the estimated category from among the plurality of past assignments;
outputting solutions associated with the searched similar past problems;
Equipped with
In the past data in which a plurality of solutions are linked to the plurality of past problems, each of the past problems belongs to one of a plurality of domains;
The classification model has been trained using training data;
The teacher data is data in which assignment tags corresponding to the plurality of past assignments are classified into the plurality of categories according to their mutual relevance,
A method for proposing a solution, in which each of the problem tags is determined based on each of the past problems and information in the past data that belongs to the domain to which each of the past problems belongs among the multiple domains and is related to each of the past problems. The past data includes hierarchical data in which problem items and means items are arranged in a hierarchical structure,
5. The method for proposing a solution according to claim 4 , wherein each of the past assignments is entered in one of the assignment items in the hierarchical data. It runs on a computer
an input unit for accepting input of a new assignment;
an estimation unit that estimates a category into which the new assignment will be classified from among a plurality of categories into which a plurality of past assignments have been classified using a classification model;
a search unit that searches for past assignments similar to the new assignment among the plurality of past assignments within the estimated category;
an output unit that outputs solutions associated with the searched similar past problems;
Equipped with
the estimation unit estimates, from among the plurality of categories, a category having a degree of relevance with the new assignment equal to or greater than a first threshold as a category into which the new assignment will be classified;
the output unit outputs the estimated degree of association between the category and the new task;
the output unit outputs a degree of relevance of a category, among the plurality of categories, whose degree of relevance with the new assignment is less than the first threshold;
The system for proposing the solution , wherein the input unit is capable of accepting a change to the first threshold value . the search unit extracts, from the plurality of past assignments, past assignments that are classified into the estimated category and have a similarity to the new assignment that is equal to or greater than a second threshold, as the similar past assignments;
The system for proposing solutions according to claim 6 , wherein the output unit outputs a similarity between the extracted similar past assignment and the new assignment. the output unit outputs a similarity of a past assignment among the plurality of past assignments whose similarity to the new assignment is less than the second threshold;
The system for proposing the solution according to claim 7 , wherein the input unit is capable of accepting a change to the second threshold value. In the past data in which a plurality of solutions are linked to the plurality of past problems, each of the past problems belongs to one of a plurality of domains;
The classification model has been trained using training data;
The teacher data is data in which assignment tags corresponding to the plurality of past assignments are classified into the plurality of categories according to their mutual relevance,
The solution proposal system according to any one of claims 6 to 8, wherein each of the assignment tags is determined based on each of the past assignments and information in the past data that belongs to the domain to which each of the past assignments belongs among the multiple domains and is related to each of the past assignments. It runs on a computer
an input unit for accepting input of a new assignment;
an estimation unit that estimates a category into which the new assignment will be classified from among a plurality of categories into which a plurality of past assignments have been classified using a classification model;
a search unit that searches for past assignments similar to the new assignment among the plurality of past assignments within the estimated category;
an output unit that outputs solutions associated with the searched similar past problems;
Equipped with
In the past data in which a plurality of solutions are linked to the plurality of past problems, each of the past problems belongs to one of a plurality of domains;
The classification model has been trained using training data;
The teacher data is data in which assignment tags corresponding to the plurality of past assignments are classified into the plurality of categories according to their mutual relevance,
A solution suggestion system in which each of the problem tags is determined based on each of the past problems and information in the past data that belongs to the domain to which each of the past problems belongs among the multiple domains and is related to each of the past problems. The past data includes hierarchical data in which problem items and means items are arranged in a hierarchical structure,
11. The system for proposing solutions according to claim 10 , wherein each of the past assignments is entered in one of the assignment items in the hierarchical data.