Information processing device and information processing method

Abstract

To provide an information processing apparatus for properly processing input data even when the input data to be processed is not similar to past input data, and an information processing method.SOLUTION: An information processing system 1 includes a first database 11, a second database 12, a third database 13, a fourth database 14, a communication device 15, a display device 16, and an information processing apparatus 17. The information processing apparatus 17 processes input data acquired from the first database 11, using learning data acquired from the second database 12, processing criterion data acquired from the third database 13, and corpus data acquired from the fourth database 14, and notifies a user of a processing result through the display device 16.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device and an information processing method for processing various types of data. [Background technology]

[0002] A method is known for predicting the stability of a power system based on input data including information on the system state estimated based on the power measured in the power system and a preset power supply and demand in the power system, and the state of an accident that is expected to occur in the power system in the estimated system state (Patent Document 1).

[0003] In the method of Patent Document 1, the degree of learning of the learning data for the input data is determined using the input data and learning data generated based on past input data using a predetermined evaluation model. If the degree of learning is high, that is, if the input data and the learning data are highly similar, the stability of the power system is estimated using a predetermined evaluation model based on the input data and the learning data, whereas if the degree of learning is low, that is, if the input data and the learning data are low similarity, the stability of the power system is calculated from the input data using a preset calculation method such as simulation. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-47696 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the above conventional technology, when the degree of learning is determined to be high, the stability of the power system can be accurately estimated, but when the degree of learning is determined to be low, the stability of the power system cannot be accurately estimated. Therefore, when the degree of learning is determined to be low and the stability of the power system is calculated by a calculation method such as a simulation, learning is performed on the evaluation model by a neural network using the input data and the calculated stability of the power system as teaching data.

[0006] However, whether the stability of the power system can be calculated from the input data using a calculation method such as a simulation depends on the type of input data. Therefore, if the learning degree is determined to be low, it is not always possible to calculate the system stability. If the stability of the power system cannot be calculated, no teaching data is available and the evaluation model cannot be changed by learning, so even if the input data is processed with the evaluation model, the stability of the power system cannot be predicted with high accuracy.

[0007] The problem to be solved by the present invention is to provide an information processing device and an information processing method that can appropriately process input data even if the input data to be processed is not similar to past input data. [Means for solving the problem]

[0008] The present invention relates to It is text data written in natural language. Enter power de Data, Processing criteria are text data that indicate classification criteria for classifying input data according to importance. Generated from data , regarding importance Rule 1 and Contains input data with associated importance Learned by machine learning using training data R, Different from Rule 1 The co-occurrence word that appears together with a word of the text data and has a frequency of occurrence equal to or higher than a predetermined value is When classifying using the first rule and the second rule, learning data including previously input data classified using the first rule and the second rule is used, and the input power de Data Among the words that appear in data Shows the frequency of occurrence of words that appear in and overlap with words First similarity and power de Data Among the words that appear in Training data and Shows the frequency of occurrence of words that appear in and overlap with words Calculate the second similarity, and if the value of the first similarity is equal to or greater than the value of the second similarity, power deAfter classifying the data according to the first rule, a part of the data classified according to the first rule is classified according to the second rule, and if the value of the first similarity is less than the value of the second similarity, the input power de The present invention solves the above problem by classifying the data according to the second rule, and then classifying a part of the data classified according to the second rule according to the first rule. Effect of the Invention

[0009] According to the present invention, even if the input data to be processed is not similar to past input data, the input data can be appropriately processed. [Brief description of the drawings]

[0010] [Figure 1] 1 is a block diagram showing an information processing system including an information processing device according to the present invention. [Diagram 2] 2 is an example of an image displayed on the display device of FIG. 1. [Diagram 3] 2 is a flowchart showing an example of a processing procedure in the information processing system of FIG. 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0012] [Information Processing System] Fig. 1 is a block diagram showing an information processing system 1 according to the present invention. The information processing system 1 shown in Fig. 1 can be used when acquiring product defect information in the market from, for example, texts posted on websites or social networking services (SNS), posts on electronic bulletin boards on an intranet within a company, and product repair reports sent to manufacturers by retailers that sold the products, and automatically classifying the acquired information according to its importance.

[0013] The information processing system 1 is a system including at least one database and an electronic computer that processes input data. As shown in Fig. 1, the information processing system 1 includes a first database 11, a second database 12, a third database 13, a fourth database 14, a communication device 15, a display device 16, and an information processing device 17. The first database 11 stores input data that is input to the information processing device 17 and processed, the second database 12 stores learning data used for machine learning executed by the information processing device 17, the third database 13 stores processing criterion data indicating a predetermined criterion for processing (e.g., classifying) the data input to the information processing device 17, and the fourth database 14 stores corpus data used in processing in the information processing device 17.

[0014] The input data stored in the first database 11 is not particularly limited as long as it is text data written in a natural language. Examples of text data include text posted to websites and SNS, posts to an intranet bulletin board, information entered into an in-house system, and product repair reports sent from a dealer that sold the product to the manufacturer. The input data in this embodiment particularly includes a quality report (technical report) sent from a dealer that sold the vehicle to the automobile company that manufactured the vehicle. The quality report includes at least information on defects in the market of the sold vehicle, such as the location where the defect occurred, the state of the defect, the cause of the defect and countermeasures, the maintenance state and repair history of the vehicle where the defect occurred, the date and time of the defect occurrence, and the date and time of the response.

[0015] The learning data stored in the second database 12 is not particularly limited as long as it can be used for machine learning executed by the information processing device 17, and is, for example, text data including documents such as laws, regulations, and guidelines related to vehicles and safety that are published by the Ministry of Land, Infrastructure, Transport and Tourism, etc. The learning data also includes input data previously processed by the information processing device 17. In this case, the learning data includes data previously input and processed by the information processing device 17 and at least the processing results at that time, and these pieces of information are associated and stored in the second database 12. For example, when input data including vehicle defect information is classified according to a classification standard described later or according to rules learned by machine learning in the information processing device 17, the input data, the classification result of the input data according to the importance of the defect, and the classification method are associated with each other and stored in the second database 12 as learning data.

[0016] The processing standard data stored in the third database 13 is data that specifies the standards used when processing input data in the information processing device 17, and is, for example, data that indicates classification standards used when classifying data input to the information processing device 17. The processing standard data is text data such as sentences written in natural language and formulas written in mathematical symbols, and examples of the data include documents related to laws and safety standards published by the Ministry of Land, Infrastructure, Transport and Tourism and the like, operating procedure manuals for classifying defects in products such as vehicles according to their severity, product cautions, and product usage manuals.

[0017] The corpus data stored in the fourth database 14 is corpus data used in processing in the information processing device 17, particularly when learning some rules by machine learning. A corpus is a structured collection of materials to be analyzed when input data is analyzed, and is a collection of written materials and recorded language materials. The type of corpus is not particularly limited, and examples include a raw corpus that is simply a collection of documents, a balanced corpus that is a balanced collection of various types of documents, a tagged corpus to which parts of speech and syntactic structures are assigned, and a parallel corpus that matches text in a certain language with its translated text. For example, if the input data is Japanese, the Balanced Corpus of Contemporary Written Japanese (BCCWJ) or the Corpus of Spoken Japanese (CSJ) can be used, and if the input data is English, the Corpus of Contemporary American English (COCA) can be used.

[0018] The first database 11 is connected to the input device 11a by known means such as wired or wireless LAN in a state in which data can be transmitted and received, and a user of the information processing system 1 inputs data to be processed by the information processing device 17 to the first database 11 using the input device 11a. The input device 11a is a device by which the user of the information processing system 1 inputs data to be processed by the information processing device 17 to the first database 11, and is, for example, a keyboard, a microphone, a touch panel, or the like connected to a display. The input device 11a may be provided together with the first database 11, or may be provided at a location away from the first database 11 and / or the information processing device 17 depending on the convenience of the user of the information processing system 1.

[0019] As shown in FIG. 1, the first database 11 is connected to a communication device 15 by a known communication means such as a wired or wireless LAN, and the information processing device 17 acquires input data stored in the first database 11 through the communication device 15. The communication device 15 is a device for transmitting and receiving data between the information processing device 17 and an external device, and is not particularly limited as long as it is a communication device capable of communicating with other devices through a network such as the Internet. Although there is one communication device 15 shown in FIG. 1, the number of communication devices 15 is not particularly limited, and multiple communication devices 15 may be used. When multiple communication devices 15 are used, the connection relationship between each device included in the information processing system 1 and the communication device 15 is not particularly limited.

[0020] 1, similarly to the first database 11, the second database 12, the third database 13, and the fourth database 14 are also connected to the communication device 15. The information processing device 17 acquires learning data from the second database 12, acquires processing reference data from the third database 13, and acquires corpus data from the fourth database 14, via the communication device 15. These databases do not need to be provided together with the information processing device 17, and at least one of the first database 11, the second database 12, the third database 13, and the fourth database 14 may be provided in a location away from the information processing device 17, for example, as the server 12a shown in FIG.

[0021] Furthermore, the information processing device 17 is also connected to the display device 16, and the processing result of the input data performed by the information processing device 17 is displayed on the display device 16. The display device 16 is a device for notifying the user of the processing status of the data input to the information processing device 17, and examples thereof include a liquid crystal display and a projector. The display device 16 in FIG. 1 is connected to the information processing device 17 in a state in which data can be exchanged, but the connection method is not particularly limited, and known communication means such as wired or wireless LAN can be used. The display device 16 may be provided together with the information processing device 17, or may be connected to the communication device 15 and provided at a location away from the information processing device 17 according to the convenience of the user of the information processing system 1.

[0022] The information processing device 17 is a device for processing input data acquired from the first database 11 using learning data acquired from the second database 12, processing reference data acquired from the third database 13, and corpus data acquired from the fourth database 14, and notifying a user of the processing result using the display device 16. The information processing device 17 realizes processing of the input data and notification of the processing result to the user by a processor 18. The processor 18 includes a ROM (Read Only Memory) 182 in which a program is stored, a CPU (Central Processing Unit) 181 which is an operating circuit for functioning as the information processing device 17 by executing the program stored in the ROM 182, and a RAM (Random Access Memory) 183 which functions as an accessible storage device.

[0023] [Input data processing section] The program used in the information processing device 17 of this embodiment includes an input data processing unit 2, which is a functional block for enabling the information processing device 17 to process input data and notify the user of the processing result. The input data processing unit 2 has a function of acquiring necessary data from each of the databases 11 to 14, classifying the input data according to a certain rule, and notifying the user of the classification result via the display device 16. As shown in Fig. 1, the input data processing unit 2 includes a data acquisition unit 21, a word occurrence extraction unit 22, a co-occurrence word extraction unit 23, a classification rule generation unit 24, a similarity calculation unit 25, an input data classification unit 26, a display image generation unit 27, and a data output unit 28. Each unit is extracted and shown in Fig. 1 for convenience.

[0024] 1 includes all the above-mentioned functional blocks, but it is not necessary for a single information processing device 17 to include all the functional blocks, and some of the above-mentioned functional blocks may be provided in other devices included in the information processing system 1, or in another information processing device not shown. For example, in the information processing system 1 of FIG. 1, the display image generating unit 27 may be provided in the display device 16. In this case, the functions of the display image generating unit 27 are executed using the CPU, ROM, and RAM of the display device 16.

[0025] In addition, it is not necessary to execute all the processes of each functional block in a single information processing device, and the functions of each functional block may be realized across a plurality of information processing devices connected in a state in which data can be exchanged. For example, in the information processing system 1 of FIG. 1, among the processes executed by the occurring word extraction unit 22, a part of the processes may be executed by the server 12a, and the remaining processes may be executed by the information processing device 17. In this case, a part of the processes for realizing the functions of the occurring word extraction unit 22 is performed using the CPU, ROM, and RAM of the server 12a. As another example, among the processes executed by each functional block of the co-occurrence word extraction unit 23 and the classification rule generation unit 24, a part of the processes may be executed by the server 12a, and the remaining processes may be executed by the information processing device 17.

[0026] The functions performed by each functional block of the input data processing unit 2 will be described below.

[0027] The data acquisition unit 21 has a function of acquiring, as necessary, the input data stored in the first database 11, the learning data stored in the second database 12, the processing reference data stored in the third database 13, and the corpus data stored in the fourth database 14. The data acquired by the data acquisition unit 21 is used in other functional blocks described below.

[0028] The word occurrence extraction unit 22 has a function of extracting words that appear in the data acquired by the data acquisition unit 21. The word occurrence extraction unit 22 has a function of extracting words used in text data included in the input data of the first database 11, the learning data of the second database 12, the processing reference data of the third database 13, and the corpus data of the fourth database 14. When the text data processed by the word occurrence extraction unit 22 is an Indo-European language in which words are written to be separated by inserting a space (blank) between each word (word segmentation), such as English, German, or French, the word occurrence extraction unit 22 extracts the words separated by the space as they are, and attaches tags such as parts of speech to each word.

[0029] The word occurrence extraction unit 22 performs stemming on the extracted words as necessary, and deletes stop words from the extracted words. Stemming is a process for regarding a certain word and its derivatives as all having the same feature. For example, for the English verb "run", if inflections such as "ran" and "running" appear in the text data, all of these inflections are considered to have the same feature as "run". A stop word is a word that appears in documents on any topic and is considered to have no relation to the type of topic. Examples include basic words such as "the", "is", "have", "take", and "get". These processes can be performed, for example, by using a module such as "termextract" in the programming language "Python".

[0030] On the other hand, when the text data to be processed by the word extraction unit 22 is a language that does not separate words, such as Japanese, Chinese, or Thai, the word extraction unit 22 performs morpheme analysis on the text data. A morpheme is the smallest linguistic unit that has meaning, and in this embodiment, morpheme analysis refers to determining word divisions in the text data and determining the part of speech and the like for each divided word. For example, a sentence such as "The engine is making a strange sound and is not working" can be divided into "The engine is making a strange sound and is not working" by morpheme analysis, and the reading, root form, part of speech, conjugation form, and the like of each divided word are determined. Stop words such as particles and auxiliary verbs may be deleted from each of these divided words as necessary. Such processing can be performed using a morpheme analysis tool such as "MeCab", "JUMAN", or "ChaSen".

[0031] Instead of morphological analysis, an n-gram technique may be used. N-gram is a technique for breaking down a character string into n-character units (n is a natural number) and analyzing the character string based on the combination of adjacent characters.

[0032] The word occurrence extraction unit 22 also has a function of searching for compound words from the extracted words. The word occurrence extraction unit 22 detects compound words that combine two to five words into one, for example. For example, when the word "exhaust / gas" is extracted, the compound word "exhaust gas" is detected, and when the word "adaptive / cruise / control" is extracted, the compound word "adaptive cruise control" (ACC) is detected.

[0033] Then, the word extraction unit 22 stores the words extracted from the input data, the words extracted from the learning data, the words extracted from the processing reference data, and the words extracted from the corpus data individually in a storage unit such as RAM 183, and outputs them to other functional blocks as necessary.

[0034] Here, when the learning data processed by the word extraction unit 22 is input data previously processed by the information processing device 17, the association between the input data and the processing result is maintained for the extracted words. For example, when the learning data is input data including vehicle defect information classified according to classification criteria described below, the association between the classification result according to the importance of the defect and the classification method is maintained for the words extracted by the word extraction unit 22. Similarly, when the processing reference data is processed by the word extraction unit 22, the association between the extracted words and the classification criteria is maintained. For example, when the processing reference data is a work procedure manual for classifying defects of a product such as a vehicle according to the importance, the association between the extracted words and the importance is maintained. Note that when already structured corpus data is used, the process of extracting words from the corpus data by the word extraction unit 22 can be omitted as necessary.

[0035] The co-occurrence word extraction unit 23 has a function of extracting words that frequently occur together with each word extracted by the occurrence word extraction unit 22. For each word extracted by the occurrence word extraction unit 22, the co-occurrence word extraction unit 23 extracts words that occur frequently together with a certain word in the same text data as words that frequently occur together with that word. The occurrence frequency is the ratio of the number of times a word that occurs together with a certain word to the total number of extracted words. In this specification, words that frequently occur together with a certain word in the same text data are also referred to as "co-occurrence words".

[0036] When co-occurrence words are extracted from the words extracted by the word occurrence extraction unit 22, for example, a co-occurrence network is generated. A co-occurrence network is a diagram that visualizes the frequency of occurrence of a certain word and its co-occurrence word in text data, the distance between them when they occur, etc. The distance between a certain word and its co-occurrence word means how many words are separated between the word and its co-occurrence word. Such a co-occurrence network can be generated using a program such as "KH Coder" or "Text Mining Studio".

[0037] When extracting co-occurrence words from the generated co-occurrence network, the co-occurrence word extraction unit 23 extracts, as co-occurrence words, words that appear together with a certain word and have a frequency of occurrence equal to or higher than a predetermined value. For example, among words that appear together with a certain word in the same text data, words that appear in the top 5 to 30 frequency of occurrence are extracted as co-occurrence words. Alternatively or in addition to this, the co-occurrence word extraction unit 23 extracts, as co-occurrence words, words that appear together with a certain word and whose distance between them is equal to or less than a predetermined distance when they appear. For example, among words that appear together with a certain word, words that are 1 to 10 words apart from each other are extracted as co-occurrence words. The predetermined value of the frequency of occurrence and the predetermined distance can be selected as appropriate values ​​according to the type and amount of data from which the word extraction unit 22 has performed word extraction processing. In addition, the number of co-occurrence words extracted by the co-occurrence word extraction unit 23 may be different for each word, and the same word may be extracted as a co-occurrence word for different words.

[0038] When the co-occurrence network is generated by the co-occurrence word extraction unit 23, information other than text data may be used. In particular, when the words extracted from the learning data by the occurrence word extraction unit 22 are associated with previous processing results, the co-occurrence word extraction unit 23 can extract co-occurrence words for each group according to the previous processing results. For example, when the words extracted from the learning data by the occurrence word extraction unit 22 are associated with classifications according to the severity of vehicle defects, the co-occurrence word extraction unit 23 can extract co-occurrence words for each classification according to the severity of the defects. Then, the co-occurrence word extraction unit 23 stores the co-occurrence words extracted from the input data, the co-occurrence words extracted from the learning data, the co-occurrence words extracted from the processing reference data, and the co-occurrence words extracted from the corpus data in a storage unit such as the RAM 183 in association with the original data processed by the occurrence word extraction unit 22, and outputs them to other functional blocks as necessary. At this time, the extracted co-occurrence words may be listed for each word.

[0039] The classification rule generation unit 24 has a function of generating rules for processing input data using at least one of the words extracted by the occurrence word extraction unit 22 and the co-occurrence words extracted by the co-occurrence word extraction unit 23. The rules for processing input data are, in particular, rules for classifying input data. The classification rule generation unit 24 expresses text data included in the input data, learning data, processing reference data, and corpus data acquired by the data acquisition unit 21 as vectors using at least one of the words extracted by the occurrence word extraction unit 22 and the co-occurrence words extracted by the co-occurrence word extraction unit 23. The text data expressed as vectors becomes the rules used when the information processing device 17 processes the input data.

[0040] The classification rule generating unit 24 can generate a plurality of rules by selecting which data to use from the input data, the learning data, the processing reference data, and the corpus data, and also selecting whether to use the words extracted by the occurrence word extracting unit 22 or the co-occurrence words extracted by the co-occurrence word extracting unit 23. In this embodiment, among these rules, the first rule is a rule in which the processing reference data is expressed as a vector using the words extracted by the occurrence word extracting unit 22, and the second rule is a rule in which the co-occurrence words of the learning data extracted by the co-occurrence word extracting unit 23 are combined and the resulting combination of co-occurrence words is expressed as a vector. In this embodiment, the learning data, which is the original data of the second rule, includes at least past input data classified by the first classification unit 26a or the second classification unit 26b described later.

[0041] The first rule is a vector representation of a predetermined rule such as a document published by a government agency, a work procedure manual, a product warning label, or a user manual, while the second rule is a vector generated by changing the learning data used. Therefore, the information processing device 17 of this embodiment processes input data, and the processed input data is associated with the processing result and fed back as learning data, thereby automatically improving the second rule and enabling the input data to be processed more appropriately using the second rule. In other words, the second rule is a rule learned by machine learning using the learning data. Hereinafter, how the first rule and the second rule are generated will be described.

[0042] The processing standard data that is the source data for the first rule are documents published by public agencies, work instructions for classifying defects of products such as vehicles according to their importance, and product warnings or usage manuals, and are predetermined rules. Examples of documents published by public agencies include automobile recall and defect information published as a database by the Ministry of Land, Infrastructure, Transport and Tourism. Examples of work instructions include standards for classifying information on vehicle defects received from dealers according to three levels of importance: high, medium, and low. Such work instructions state, for example, that if the reported defect corresponds to "all the doors of the vehicle cannot be opened from the inside," the defect is classified as "high importance," if the defect corresponds to "some doors of the vehicle cannot be opened or unlocked from the inside," the defect is classified as "medium importance," and if the defect corresponds to "some doors of the vehicle cannot be opened or unlocked from the outside," the defect is classified as "low importance." For example, if the product is a hairspray that uses gas, a warning or instruction manual for the product would be something like "Do not place the product near open flames as it may explode or be damaged" written on the spray can.

[0043] In order to generate the first rule, the classification rule generating unit 24 obtains the words extracted from the processing reference data as described above from the word occurrence extracting unit 22. In this embodiment, it is assumed that the stemming process, the stop word removal, and the compound word search are performed in the word occurrence extracting unit 22 at this stage. The extracted words are listed for each text data or each sentence. Then, the extracted words are arranged horizontally as word labels, and for each text data or sentence, a one-row matrix is ​​created by assigning, for example, 1 to the word that appears and 0 to the word that does not appear. This matrix with the word labels is also called a word vector. Then, this process is performed for each text data or sentence, and the generated word vectors are collected into one to become the first rule.

[0044] For example, in the above-mentioned operating procedure manual, for the phrase "All the doors of the vehicle do not open from the inside" classified as "high importance", the morphological analysis function of the word extraction unit 22 can extract words by dividing it into "vehicle / all / of / the / doors / do / not / open / from / the inside". From these extracted words, the particles "no", "ga" and "kara" are deleted as stop words. Although "nai" is an auxiliary verb and may be a stop word, it is not deleted and "akaka (original form: akaru) / nai" is extracted as the compound word "akanai". As a result, the word "vehicle / all / doors / inside / do not open" can be extracted from the processing reference data "All the doors of the vehicle do not open from the inside". Then, from these extracted words, a word vector shown in the "high" row of Table 1 below is generated for classifying the input data as "high importance".

[0045] Similarly, for "some of the doors of the vehicle cannot be opened or unlocked from the inside" classified as "medium importance", the morphological analysis function of the word extraction unit 22 extracts words by dividing them into "some of the doors of the vehicle cannot be opened from the inside, or cannot be unlocked", removes "of", "ga", "kara", "," and "or" as stop words, and extracts "cannot open" as the compound word "cannot open" and "cannot unlock" as the compound word "cannot unlock". As a result, the word "vehicle / some / door / inside / cannot open / cannot unlock" can be extracted from the processing reference data "some of the doors of the vehicle cannot be opened or unlocked from the inside". Then, from these extracted words, a word vector shown in the "medium" row of Table 1 below is generated for classifying the input data as "medium importance".

[0046] Similarly, for "Some doors of the vehicle cannot be opened from the outside or cannot be unlocked" classified as "low importance", the morphological analysis function of the word extraction unit 22 extracts words by dividing them into "Some doors of the vehicle cannot be opened from the outside, or cannot be unlocked", removes "of", "ga", "kara", "," and "or" as stop words, and extracts "cannot open" as the compound word "cannot open" and "cannot unlock" as the compound word "cannot unlock". As a result, the word "vehicle / some / doors / outside / cannot open / cannot unlock" can be extracted from the processing reference data "Some doors of the vehicle cannot be opened from the outside or cannot be unlocked". Then, from these extracted words, a word vector shown in the "low" row of Table 1 below is generated for classifying the input data as "low importance".

[0047] [Table 1]

[0048] In this way, even when the processing reference data is processed by the word occurrence extraction unit 22, the association between the extracted words and the classification references is maintained, so that each word vector can be associated with the classification references.

[0049] On the other hand, in order to generate the second rule, the classification rule generating unit 24 acquires co-occurring words extracted from the learning data from the co-occurring word extracting unit 23. When generating the second rule, the classification rule generating unit 24 extracts combinations of words with strong co-occurrence relationships from the acquired list of co-occurring words. For example, the classification rule generating unit 24 extracts combinations of words that are in the top 5 most frequently occurring words among words that appear together with a certain word in the same text data, and that are 5 words or less away from each other. The conditions for the extraction can be set appropriately within a range in which the number of combinations to be extracted does not exceed the processing limit of the information processing device 17. Then, the extracted words are arranged horizontally as word labels, and 1 is assigned to the words extracted as the combination, and 0 is assigned to the other words, to create a word vector for each combination. This process is performed for each combination, and the generated word vectors are collected into one to become the second rule.

[0050] In this embodiment, the extracted co-occurrence words are listed for each word, and the acquired co-occurrence words are extracted for each group according to the previous processing result. By extracting the co-occurrence words for each group according to the previous processing result, each word vector can be associated with the processing result. For example, if the co-occurrence words used to obtain the second rule are extracted for each group according to the severity of the vehicle defect, the generated word vector can be associated with the severity of the defect. In other words, when the input data includes information on the vehicle defect, the second rule can be used as a rule for classifying the data according to the severity of the vehicle defect.

[0051] For example, when the learning data includes data that was previously input to the information processing device 17 and classified as "high importance" by the first classification unit or the second classification unit, the classification rule generation unit 24 acquires co-occurrence words of the data classified as "high importance" from the co-occurrence words extracted by the co-occurrence word extraction unit 23. When the co-occurrence words of the data classified as "high importance" are (door, front, right side, left side, strange sound, all, does not open), the classification rule generation unit 24 extracts "door, strange sound, all, does not open" as an example of a combination of co-occurrence words. Then, the words included in this combination are arranged horizontally as word labels, and a one-row matrix is ​​created by assigning 1 to the words included in the combination and 0 to the other words. If the combination is "door, front, strange sound, all, does not open", a one-row matrix shown in the "high" row of Table 2 below, that is, a word vector, is generated. The input data is classified as "high importance" by this word vector.

[0052] Similarly, when data previously classified as "medium importance" by the first classification unit or the second classification unit is included in the learning data, the classification rule generation unit 24 acquires co-occurrence words of the data classified as "medium importance" from the co-occurrence words extracted by the co-occurrence word extraction unit 23. When the co-occurrence words of the data classified as "medium importance" are (vehicle, both sides, door, strange sound, behind, does not open), the classification rule generation unit 24 extracts "vehicle, door, strange sound, behind, does not open" as an example of a combination of co-occurrence words. Then, from this combination, a word vector shown in the "medium" row of Table 2 below is generated. The input data is classified as "medium importance" based on this word vector.

[0053] Similarly, when data previously classified as "low importance" by the first classification unit or the second classification unit is included in the learning data, the classification rule generation unit 24 acquires co-occurrence words of the data classified as "low importance" from the co-occurrence words extracted by the co-occurrence word extraction unit 23. When the co-occurrence words of the data classified as "low importance" are (vehicle, door, difficult to open, front, strange sound, rear), the classification rule generation unit 24 extracts "door, difficult to open, strange sound, rear" as an example of a combination of co-occurrence words. Then, from this combination, a word vector shown in the "low" row of Table 2 below is generated. The input data is classified as "low importance" based on this word vector.

[0054] [Table 2]

[0055] In this way, co-occurring words are extracted for each classification result of the learning data, so that each word vector can be associated with a classification criterion. Here, a process has been described in which the first and second rules are used to classify data according to the severity of a vehicle malfunction when the input data includes information about a vehicle malfunction, but the classification criterion is not limited to the severity of the vehicle malfunction, and may be the location where the vehicle malfunction occurred, the maintenance status of the vehicle where the malfunction occurred, etc. Also, the method of generating the first and second rules described above is merely an example, and other methods may be used.

[0056] The similarity calculation unit 25 has a function of calculating the similarity between the data input to the information processing device 17 and the data on which the rule (i.e., one or more word vectors) generated by the classification rule generation unit 24 is based. In particular, the similarity calculation unit 25 of this embodiment includes a first similarity calculation unit that calculates a first similarity, which is the similarity between the input data and the predetermined data on which the first rule is based (i.e., processing reference data), and a second similarity calculation unit that calculates a second similarity, which is the similarity between the input data and the learning data. In the calculation of the similarity in the first similarity calculation unit and the second similarity calculation unit, the word vector generated from the data input from the input device 11a can be used. The method of generating the word vector from the input data is the same as the above-mentioned method of generating the first rule. For example, if the input data is "The front door of the vehicle cannot be opened from the inside" and "All the doors cannot be opened from the outside", the word vectors in Table 3 below are generated through morphological analysis, removal of stop words, and extraction of compound words.

[0057] [Table 3]

[0058] The similarity calculated by the similarity calculation unit 25 is an index showing the degree to which a certain text data overlaps with another text data, and takes a positive value within the range of 0 to 1, for example. The higher the similarity value, the more similar the text data to be compared are, and the lower the similarity value, the more dissimilar the text data to be compared are. For example, the frequency of occurrence of words that overlap with words that appear in a predetermined processing reference data among words that appear in data input from the input device 11a is calculated as the first similarity, and the frequency of occurrence of words that overlap with words that appear in the learning data among words that appear in data input from the input device 11a is calculated as the second similarity. The frequency of occurrence of such overlapping words can be calculated by comparing a word vector generated from the input data with a word vector constituting the first or second rule. Specifically, the word label of the word vector generated from the input data is compared with the word label attached to the first or second rule to extract overlapping words. Then, the frequency of occurrence of the extracted overlapping words is calculated as the similarity.

[0059] Alternatively, the word vector generated from the input data may be compared with the word vector of the first or second rule, and the similarity may be calculated for each word vector. Specifically, the word vector generated from the input data may be compared with the word vector of the first or second rule, and the ratio of the number of matching word vector values ​​(i.e., 0 or 1) to the total number of word labels may be calculated as the similarity. Furthermore, the average value of the similarities for each word vector may be calculated as the first similarity or the second similarity.

[0060] Furthermore, when calculating the similarity taking into account the meaning of a word, co-occurring words of the words are compared. In other words, when the meaning of a word is similar to that of another word, the co-occurring words of the word are also similar to those of the other words. In this embodiment, by using a co-occurring word list extracted from corpus data as an index, it is possible to compare the co-occurring words of a word in the corpus data with the co-occurring words of the other words and calculate the similarity of the meaning between the words. For example, the Simpson-Jaccard coefficient can be used to compare the co-occurring words. The Simpson-Jaccard coefficient is an index obtained by arithmetically averaging the Jaccard coefficient and the Simpson coefficient, and is expressed by the following formula (1).

[0061]

number

[0062] Here, w1 and w2 in equation (1) represent words, Jaccard(w1,w2) is the Jaccard coefficient, and Simpson(w1,w2) is the Simpson coefficient.

[0063] Specifically, the word labels of the word vectors generated from the input data are compared with the word labels of the first or second rule to extract overlapping words. Next, for the extracted overlapping words, a co-occurrence word list extracted from corpus data is obtained. The co-occurrence word list is generated by the co-occurrence word extraction unit 23. Then, the Simpson-Jaccard coefficient is calculated using the overlapping words and the co-occurrence word list extracted from the corpus data.

[0064] As an example, the similarity between the word label [engine, strange sound, doesn't work] of the word vector generated from the input data and the word label [engine, strange sound, doesn't work] of the first or second rule is calculated as the Simpson-Jaccard coefficient. If it is known from the co-occurrence word list extracted from the corpus data that the similarity between [strange sound] and [strange sound] is 0.75, and the similarity between [doesn't work] and [doesn't work] is 0.65, the Simpson-Jaccard coefficient in this case is (1.0+0.75+0.65) / 3=0.8. Here, since [engine] and [engine] are completely identical, the similarity is calculated as 1. Also, the similarity may be calculated using cosine similarity instead of the Simpson-Jaccard coefficient.

[0065] The input data classification unit 26 has a function of classifying data input by the input device 11a according to a rule. In particular, the input data classification unit 26 of this embodiment includes a first classification unit 26a that classifies input data according to a first rule, and a second classification unit 26b that classifies input data according to a second rule. As described above, in the first rule, a word vector is associated with a classification criterion, and in the second rule, a word vector is associated with a previous processing result. Therefore, the first classification unit 26a compares the word vector of the input data with the word vector of the first rule, and if the similarity is equal to or greater than a predetermined value, the input data can be classified according to the associated classification criterion. Similarly, the second classification unit 26b compares the word vector of the input data with the word vector of the second rule, and if the similarity is equal to or greater than a predetermined value, the input data can be classified according to the associated previous processing result.

[0066] In the input data classification unit 26 of this embodiment, when the value of the similarity calculated by the first similarity calculation unit 25a is equal to or greater than the value of the similarity calculated by the second similarity calculation unit 25b, the input data is classified by the first classification unit 26a, and then a part of the data classified by the first classification unit 26a is classified by the second classification unit 26b. On the other hand, when the value of the similarity calculated by the first similarity calculation unit 25a is less than the value of the similarity calculated by the second similarity calculation unit 25b, the input data is classified by the second classification unit 26b, and then a part of the data classified by the second classification unit 26b is classified by the first classification unit 26a. This simplifies the classification process for characteristic data (e.g., data with high importance) that does not need to be classified twice by the first and second rules, and the processing time in the information processing device 17 can be shortened.

[0067] For example, when the input data includes information on a vehicle malfunction and the input data is classified into three or more groups (for example, three groups of high importance, medium importance, and low importance) according to the importance of the vehicle malfunction, data classified as high or medium importance by one of the first classification unit 26a and the second classification unit 26b does not need to be classified by the other of the first classification unit 26a and the second classification unit 26b. In this case, the classification process is performed only on data classified as low importance by the other of the first classification unit 26a and the second classification unit 26b.

[0068] In the input data classification unit 26 of the present embodiment, the first similarity value and the second similarity value are compared, and if the first similarity value is equal to or greater than the second similarity value, the input data is classified in the order of the first classification unit 26a → the second classification unit 26b, and if the first similarity value is less than the second similarity value, the input data is classified in the order of the second classification unit 26b → the first classification unit 26a. The calculation of the first similarity is not essential to the present invention and can be omitted as necessary. For example, the second similarity value may be calculated, and if the second similarity value is equal to or greater than a predetermined value, the input data is classified in the order of the first classification unit 26a → the second classification unit 26b, and if the second similarity value is less than the predetermined value, the input data is classified in the order of the second classification unit 26b → the first classification unit 26a. The predetermined value of the similarity can be set to an appropriate value within a range in which the data input to the information processing device 17 is appropriately processed.

[0069] The display image generating unit 27 has a function of generating an image to be displayed on the display device 16 to notify the user of the processing status of the input data. The image generated by the display image generating unit 27 may indicate the progress of the processing or the processing result, and may include, for example, at least one of the result of classification by the first classification unit, the result of classification by the second classification unit, the similarity calculated by the first similarity calculation unit, and the similarity calculated by the second similarity calculation unit.

[0070] Fig. 2 is a diagram showing an example of an image displayed on the display device 16 of Fig. 1. The image shown in Fig. 2 is an image displayed on the display device 16 when input data, including information on a vehicle malfunction, is classified into three groups, i.e., high importance, medium importance, and low importance, according to the importance of the vehicle malfunction. Fig. 2(A) is a first display image 3 showing an overview of the classification process, and Fig. 2(B) is a second display image 4 showing the details of the classification process.

[0071] The first display image 3 in FIG. 2(A) shows the importance of the defect as determined as a result of the classification process, as shown in the "Importance Determination Result" displayed in the upper left of the image. On the left side of the first display image 3, "1000 items" is displayed as the "Number of Input Data", and the total number of data items input to the information processing device 17 is displayed. Below that, the number of input data items that fall into each of the high, medium, and low importance groups is displayed. In the image of FIG. 2(A), this shows that "300 items" of data fall into the "high importance" group, "300 items" of data fall into the "medium importance" group, and "400 items" of data fall into the "low importance" group.

[0072] On the other hand, in the upper right corner of the first display image 3, "B. Machine learning model → A. Rule-based model" is displayed as a "judgment method". This "judgment method" indicates whether the first classification unit 26a or the second classification unit 26b processes the data first. In the case of "B. Machine learning model → A. Rule-based model", it indicates that the second classification unit 26b classifies the data according to the second rule learned by machine learning, and then the first classification unit 26a classifies the data according to a predetermined first rule. In contrast, in the case of "A. Rule-based model → B. Machine learning model", it indicates that the first classification unit 26a classifies the data according to the first rule, and then the second classification unit 26b classifies the data according to the second rule. In addition, an image switching button 31 is displayed in the lower right corner of the first display image 3. If the display device 16 is a touch panel, touching the image switching button 31 switches the display image from the first display image 3 to the second display image 4.

[0073] In the second display image 4 in FIG. 2(B), "Determination method" is displayed at the top of the image, and information on which the data is classified in the order of "B. Machine learning model → A. Rule-based model" is displayed. In the second display image 4, columns are provided to display word labels of "Learning data", "Input data", and "Work procedure manual" from the left. Here, the work procedure manual is the original data used when generating the predetermined first rule. In the center column, the word labels extracted by the appearing word extraction unit 22 for the input data are displayed for each word vector. The list of the word labels is intended to display about the top 30, and all the word labels can be confirmed by scrolling the display portion.

[0074] In the left column of the input data, the word labels of the second rule extracted by the classification rule generating unit 24 for the used learning data are displayed for each word vector. In addition, the similarity with the input data is displayed for each word vector, and the maximum value of the similarities calculated as the second similarity is displayed at the top of the input data column. In the case of FIG. 2(B), the maximum value of the second similarity is 0.9. On the other hand, in the right column of the input data, the word labels of the first rule extracted by the classification rule generating unit 24 for the work procedure manual are displayed for each word vector. In addition, the similarity with the input data is displayed for each word vector, and the maximum value of the similarities calculated as the first similarity is displayed at the top of the input data column. In the case of FIG. 2(B), the maximum value of the first similarity is 0.8. Therefore, in the case of FIG. 2, since the value of the second similarity is larger than the value of the first similarity, it can be seen that the input data is processed first by the second classification unit 26b.

[0075] The data output unit 28 has a function of outputting the image data generated by the display image generation unit 27 to the display device 16. This allows an image to be displayed on the display device 16, making it possible to notify the user of the information processing system 1 of the processing status of the input data.

[0076] [Processing in information processing systems] The procedure for information processing device 17 to process information will be described with reference to Fig. 3. Fig. 3 is an example of a flowchart showing information processing in information processing system 1 of the present embodiment. The process described below is executed by processor 18 of information processing device 17 at predetermined time intervals.

[0077] First, in step S1, the data acquisition unit 21 acquires necessary data from each of the first database 11 to the fourth database 14. In the following step S2, the occurrence word extraction unit 22 extracts words used in the data acquired by the data acquisition unit 21 using a morphological analysis tool or the like. For example, from the data "all doors do not open from the inside", the words "vehicle / all / doors / inside / do not open" are extracted through morphological analysis, stop word deletion, and compound word extraction. In the following step S3, the co-occurrence word extraction unit 23 generates a co-occurrence network and extracts co-occurrence words, particularly for words extracted from the learning data, the processing reference data, and the corpus data. In the following step S4, the classification rule generation unit 24 generates a first rule using the words extracted from the processing reference data, and generates a second rule using the co-occurrence words extracted from the learning data.

[0078] Next, in step S5, the similarity between the input data and the word vector of the rule is calculated by the function of the similarity calculation unit 25. Specifically, the first similarity calculation unit 25a calculates a first similarity which is the similarity between the input data and the predetermined processing reference data on which the first rule is based, and the second similarity calculation unit 25b calculates a second similarity which is the similarity between the input data and the learning data. In the following step S6, the first similarity and the second similarity are compared to determine whether the value of the first similarity is equal to or greater than the value of the second similarity.

[0079] If it is determined that the value of the first similarity is equal to or greater than the value of the second similarity, the process proceeds to step S7a. In step S7a, the first classification unit 26a classifies the input data according to the first rule, and in the subsequent step S8a, a portion of the data classified by the first classification unit 26a is classified by the second classification unit 26b. On the other hand, if it is determined that the value of the first similarity is less than the value of the second similarity, the process proceeds to step S7b. In step S7b, the second classification unit 26b classifies the input data according to the second rule, and in the subsequent step S8b, a portion of the data classified by the second classification unit 26b is classified by the first classification unit 26a.

[0080] The results of these classification processes are processed in the following step S9, and an image to be displayed on the display device 16 is generated by the function of the display image generation unit 27. The generated image is output to the display device 16 by the function of the data output unit 28 in the following step S10. Then, in step S11, the image output from the data output unit 28 is displayed on the display device 16, and the user of the information processing system 1 is notified of the processing result.

[0081] [Embodiments of the present invention] As described above, according to the information processing device 17 of the present embodiment, the information processing device 17 includes a first classification unit 26a that classifies input data according to a first rule generated from predetermined data, a second classification unit 26b that classifies the input data according to a second rule different from the first rule that is learned by machine learning using learning data, a first similarity calculation unit 25a that calculates a similarity between the input data and the predetermined data, and a second similarity calculation unit 25b that calculates a similarity between the input data and the learning data. In the information processing device 17, the learning data is input in the past and classified into the first classification unit 26a or the second classification unit 26b. When the value of the similarity calculated by the first similarity calculation unit 25a is equal to or greater than the value of the similarity calculated by the second similarity calculation unit 25b, the input data is classified by the first classification unit 26a, and then a part of the data classified by the first classification unit 26a is classified by the second classification unit 26b. When the value of the similarity calculated by the first similarity calculation unit 25a is less than the value of the similarity calculated by the second similarity calculation unit 25b, the input data is classified by the second classification unit 26b, and then a part of the data classified by the second classification unit 26b is classified by the first classification unit 26a. This makes it possible to appropriately process the input data even if the input data to be processed is not similar to past input data. In addition, it is possible to simplify the classification process for characteristic data (e.g., data with high importance) that does not need to be classified twice by the first and second rules, and it is possible to shorten the processing time in the information processing device 17.

[0082] Furthermore, according to the information processing device 17 of this embodiment, at least one of the input data and the learning data includes text data written in a natural language, which makes it possible to process data that cannot be directly processed by a calculation method such as a simulation.

[0083] According to the information processing device 17 of the present embodiment, the first similarity calculation unit 25a calculates the frequency of occurrence of words that overlap with words that appear in the predetermined data among the words that appear in the input data as the similarity. This makes it possible to suppress the amount of data to be classified according to the second rule, and shorten the processing time in the information processing device 17.

[0084] According to the information processing device 17 of the present embodiment, the second similarity calculation unit 25b calculates the frequency of occurrence of words that overlap with words that appear in the learning data among the words that appear in the input data as the similarity. This makes it possible to suppress the amount of data to be classified according to the first rule, and shorten the processing time in the information processing device 17.

[0085] According to the information processing device 17 of the present embodiment, the first similarity calculation unit 25a and the second similarity calculation unit 25b each calculate the similarity using a word vector generated from the input data, thereby making it possible to process data in a natural language that cannot be directly processed by a calculation method such as a simulation.

[0086] Moreover, according to the information processing device 17 of the present embodiment, each of the first similarity calculation unit 25a and the second similarity calculation unit 25b calculates the similarity using the Simpson-Jaccard coefficient, which allows the similarity to be calculated more accurately.

[0087] Moreover, according to the information processing device 17 of this embodiment, the second rule is a rule learned by machine learning using a co-occurrence network. This makes it possible to learn the second rule more effectively.

[0088] Furthermore, according to the information processing device 17 of this embodiment, the first classification unit and the second classification unit classify the input data into at least three groups, thereby making it possible to realize classification in accordance with the input data.

[0089] According to the information processing device 17 of the present embodiment, the input data includes information on vehicle malfunctions, and the first rule and the second rule are rules for classifying the malfunctions according to their severity. This makes it possible to classify the input data according to the severity of the malfunctions.

[0090] Moreover, the information processing device 17 of this embodiment further includes a display device 16 that displays at least one of the result of classification by the first classification unit 26a, the result of classification by the second classification unit 26b, the similarity calculated by the first similarity calculation unit 25a, and the similarity calculated by the second similarity calculation unit 25b. This makes it possible to notify the user of the information processing system 1 of the processing status of the input data.

[0091] According to the information processing method of the present embodiment, when input data is classified using a first rule generated from predetermined data and a second rule different from the first rule, which is learned by machine learning using learning data, the learning data includes previously input data classified using the first rule and the second rule, a first similarity between the input data and the predetermined data is calculated, a second similarity between the input data and the learning data is calculated, and if the value of the first similarity is equal to or greater than the value of the second similarity, the input data is classified according to the first rule, and then a part of the data classified according to the first rule is classified according to the second rule, and if the value of the first similarity is less than the value of the second similarity, the input data is classified according to the second rule, and then a part of the data classified according to the second rule is classified according to the first rule. This makes it possible to appropriately process input data even if the input data to be processed is not similar to past input data. In addition, the classification process can be simplified for characteristic data (for example, data of high importance) that does not need to be classified twice according to the first and second rules, and the time required for the process can be shortened. [Explanation of symbols]

[0092] 1. Information processing system 11...First database (input data) 11a…Input device 12…Second database (learning data) 12a…Server 13...Third database (processing standard data) 14…Fourth database (corpus data) 15...Communication equipment 16...Display device 17...Information processing device 18…Processor 181…CPU 182...ROM 183…RAM 2...Input data processing section 21…Data acquisition section 22... Word extraction section 23...Collocation word extraction section 24...Classification rule generation unit 25…Similarity calculation unit 25a...first similarity calculation unit 25b…Second similarity calculation unit 26...Input data classification section 26a...First classification part 26b…Second classification part 27...Display image generating unit 28...Data output section 3…First display image 31...Image switching button 4...Second display image

Claims

1. A first classification unit that classifies input data, which is text data written in a natural language, according to a first rule regarding the importance generated from processing reference data, which is text data indicating classification criteria for classifying the input data according to importance; a second classification unit that is trained by machine learning using training data including the input data associated with the importance, and that classifies the input data according to a second rule, which is different from the first rule and is related to a co-occurring word that appears together with a word of the text data at a frequency equal to or higher than a predetermined value; a first similarity calculation unit that calculates a first similarity indicating an appearance frequency of a word that appears in the input data and that overlaps with a word that appears in the processing reference data; a second similarity calculation unit that calculates a second similarity indicating an appearance frequency of a word that appears in the input data and that overlaps with a word that appears in the learning data, The learning data includes data that was previously input and classified by the first classification unit or the second classification unit, when the value of the first similarity calculated by the first similarity calculation unit is equal to or greater than the value of the second similarity calculated by the second similarity calculation unit, after classifying the input data by the first classification unit, classifying a part of the data classified by the first classification unit by the second classification unit; an information processing device which, when the first similarity value calculated by the first similarity calculation unit is less than the second similarity value calculated by the second similarity calculation unit, classifies the input data by the second classification unit, and then classifies a portion of the data classified by the second classification unit by the first classification unit.

2. 2. The information processing device according to claim 1, wherein the first similarity calculation unit and the second similarity calculation unit respectively calculate the first similarity and the second similarity using word vectors generated from the input data and indicating words that appear and words that do not appear in a sentence of the text data of the input data.

3. 3 . The information processing apparatus according to claim 1 , wherein the first similarity calculation unit and the second similarity calculation unit calculate the first similarity and the second similarity using a Simpson-Jaccard coefficient, respectively.

4. The information processing device according to claim 1 , wherein the second rule is a rule learned by machine learning using a co-occurrence network.

5. 5. The information processing apparatus according to claim 1, wherein the first classification unit and the second classification unit classify the input data into at least three groups.

6. The input data includes information on vehicle malfunctions, 6. The information processing apparatus according to claim 1, wherein the first rule and the second rule are rules for classifying the defects according to the importance of the defects.

7. The information processing device according to any one of claims 1 to 6, further comprising a display device that displays at least one of the result of classification by the first classification unit, the result of classification by the second classification unit, the first similarity calculated by the first similarity calculation unit, and the second similarity calculated by the second similarity calculation unit.

8. An information processing method executed by an information processing device which classifies input data, which is text data written in a natural language, using a first rule regarding the importance generated from processing standard data, which is text data indicating classification standards for classifying the input data according to importance, and a second rule regarding co-occurring words that appear together with a word in the text data and have a frequency of occurrence equal to or higher than a predetermined value, the second rule being learned by machine learning using training data including the input data with which the importance is associated, the second rule being different from the first rule the learning data includes previously input data classified using the first rule and the second rule; The information processing device includes: calculating a first similarity indicating an appearance frequency of words appearing in the input data that overlap with words appearing in the processing reference data; calculating a second similarity indicating an occurrence frequency of words that appear in the input data and that overlap with words that appear in the learning data; if the value of the first similarity is equal to or greater than the value of the second similarity, classifying the input data according to the first rule, and then classifying a portion of the data classified according to the first rule according to the second rule; An information processing method comprising: when the first similarity value is less than the second similarity value, classifying the input data according to the second rule, and then classifying a portion of the data classified according to the second rule according to the first rule.

Images