Natural language processing system using dynamics calculation, natural language processing method and program

Abstract

To quantitatively evaluate a natural sentence.SOLUTION: A natural language processing system comprises: a coordinate setting part 23a that for each of a plurality of elements constituting a natural sentence to be analyzed, sets virtual coordinates of the elements based on the distributional representation of the elements; a mass setting part 25 that considers the plurality of elements as a plurality of virtual particles placed on the virtual coordinates corresponding to each element, and sets a virtual mass to the plurality of virtual particles; a displacement calculation part 27 that by applying a potential function to the plurality of virtual particles to perform dynamics calculation, calculates displacement of the plurality of virtual particles; and an evaluation value calculation part 28 that on the basis of the displacement calculated for the plurality of virtual particles, calculates an evaluation value indicating an interaction between the plurality of elements constituting the natural sentence to be analyzed.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to techniques for processing natural language. [Background technology]

[0002] Conventionally, a patent information analysis device has been proposed which comprises: means for searching a set of patent documents based on a predetermined viewpoint; means for extracting a plurality of classification codes assigned to each of the patent documents belonging to the set obtained by the search to create a set of classification codes; means for selecting a classification code for numerical analysis from the set of classification codes; means for calculating the coordinates of the classification code for numerical analysis by numerical analysis; and means for calculating the coordinates of each of the patent documents based on the coordinates of the classification code for numerical analysis, and creating a map which represents the density of the patent documents based on their coordinates (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-207173 A Summary of the Invention [Problem to be solved by the invention]

[0004] Various techniques for evaluating natural language sentences have been proposed in the past. However, the conventional techniques have room for improvement in terms of quantitatively evaluating natural language sentences.

[0005] In view of the above problems, an object of the present disclosure is to quantitatively evaluate natural language. [Means for solving the problem]

[0006] One example of the present disclosure is a natural language processing system that uses dynamics calculations, including: a coordinate setting means that sets virtual coordinates of a plurality of elements constituting a natural sentence to be analyzed based on a distributed representation of the element, for each of the plurality of elements, a mass setting means that regards the plurality of elements as a plurality of virtual particles placed at the virtual coordinates corresponding to each element and sets virtual masses to the plurality of virtual particles, a displacement calculation means that calculates displacements of the plurality of virtual particles by applying a potential function to the plurality of virtual particles to perform dynamics calculations, and an evaluation value calculation means that calculates an evaluation value indicative of an interaction between the plurality of elements constituting the natural sentence to be analyzed based on the displacements calculated for the plurality of virtual particles.

[0007] Moreover, one example of the present disclosure includes a unit division means for dividing a natural sentence to be analyzed into a plurality of sentence units, an element coordinate setting means for setting, for each of a plurality of elements constituting each of the plurality of sentence units, virtual coordinates of the virtual particles when the element is regarded as a virtual particle based on a distributed representation of the element corresponding to the virtual particle, an inter-element energy calculation means for calculating, for each of the plurality of sentence units, energy indicating an interaction between the virtual particles corresponding to the plurality of elements by regarding the plurality of elements constituting the target sentence unit as a plurality of virtual particles placed at the virtual coordinates corresponding to each element and applying a potential function to the plurality of virtual particles to perform a force field calculation, and a unit coordinate setting means for setting virtual coordinates of the sentence unit; a mass setting means for regarding the sentence units as a plurality of virtual particles placed at the virtual coordinates corresponding to each sentence unit and for setting a virtual mass for each of the plurality of sentence units based on the energy indicating the interaction between the plurality of virtual particles corresponding to the plurality of elements constituting a target sentence unit; a displacement calculation means for calculating displacements of the plurality of virtual particles corresponding to the plurality of sentence units by performing dynamics calculations by applying a potential function to the plurality of virtual particles corresponding to the plurality of sentence units; and an evaluation value calculation means for calculating an evaluation value indicating the interaction between the plurality of sentence units constituting the natural sentence to be analyzed based on the displacements calculated for the plurality of virtual particles.

[0008] The present disclosure can be understood as an information processing device, a system, a method executed by a computer, or a program executed by a computer. The present disclosure can also be understood as such a program recorded on a recording medium readable by a computer or other device, machine, etc. Here, a recording medium readable by a computer, etc. refers to a recording medium that stores information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer, etc. Effect of the Invention

[0009] According to the present disclosure, it becomes possible to quantitatively evaluate natural language sentences. [Brief description of the drawings]

[0010] [Figure 1] 1 is a schematic diagram showing a configuration of an information processing system according to a first embodiment. [Diagram 2] 1 is a diagram illustrating an outline of a functional configuration of an information processing device according to a first embodiment. [Diagram 3] 1 is a diagram showing a concept in which elements constituting natural language are regarded as virtual particles and displacements after dynamics calculation are calculated in the first embodiment. FIG. [Figure 4] 4 is a flowchart showing the flow of a natural language analysis process according to the first embodiment. [Diagram 5] 3 is a schematic diagram showing a state in which a plurality of elements to which virtual coordinates are assigned are plotted in a three-dimensional space in the first embodiment. FIG. [Figure 6] 4 is a diagram showing a concept of setting a virtual mass to a virtual particle corresponding to an element included in a natural sentence to be analyzed in the first embodiment. FIG. [Figure 7] FIG. 2 is a schematic diagram showing how dynamics calculation is performed by applying a potential function between virtual particles corresponding to a plurality of elements constituting a natural language in the first embodiment. [Figure 8] FIG. 4 is a diagram showing a concept of performing evaluation according to a displacement amount in the first embodiment. [Figure 9] FIG. 2 is a diagram illustrating an overview of a process for acquiring attributes of elements of natural language using a machine learning model in the first embodiment. [Figure 10] FIG. 11 is a diagram showing an overview of the relationship between the distance between virtual particles and energy in the case where a function for calculating an interatomic potential is adopted as the potential function in the first embodiment. [Figure 11]FIG. 11 is a diagram showing an example of parameters set in a potential function when a function for calculating an interatomic potential is adopted as the potential function in the first embodiment. [Figure 12] FIG. 11 is a diagram illustrating an outline of a functional configuration of an information processing device according to a second embodiment. [Figure 13] 13 is a flowchart showing the flow of a natural language analysis process according to a second embodiment. [Figure 14] FIG. 13 is a diagram illustrating an outline of the functional configuration of an information processing device according to a third embodiment. [Figure 15] FIG. 13 is a diagram showing the concept in which elements and sentence units constituting natural language sentences are regarded as virtual particles and energy is calculated in the third embodiment. [Figure 16] 13 is a flowchart showing the flow of a natural language analysis process according to the third embodiment. [Figure 17] FIG. 11 is a schematic diagram showing how a sentence unit including a plurality of elements is regarded as a virtual particle including virtual particles corresponding to these sentence units in the third embodiment. [Figure 18] FIG. 13 is a diagram illustrating an outline of a functional configuration of an information processing device according to a fourth embodiment. [Figure 19] 13 is a flowchart showing the flow of a natural language analysis process according to the fourth embodiment. [Figure 20] FIG. 13 is a diagram illustrating an example of an analysis of a case-dominating relationship in the fourth embodiment. [Figure 21] FIG. 13 is a diagram showing the concept of dynamics calculation in a state in which some virtual particles are fixed in the fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] <First embodiment> Hereinafter, an embodiment of an information processing device, method, and program according to the present disclosure will be described with reference to the drawings. However, the embodiment described below is merely an example of an embodiment, and the information processing device, method, and program according to the present disclosure are not limited to the specific configuration described below. In carrying out the present invention, a specific configuration according to the embodiment may be appropriately adopted, and various improvements and modifications may be made. The present invention can appropriately adopt at least a part of the configurations in the embodiments and variations described below.

[0012] In this embodiment, the technology according to the present disclosure is described in terms of its implementation in a patent value assessment system or patent information analysis system for applying dynamics calculations to the evaluation of patent claims (referred to as "patent dynamics" herein). However, the technology according to the present disclosure can be widely used in technologies for analyzing natural language, and the application of the present disclosure is not limited to the examples shown in the embodiment. For example, the technology according to the present disclosure may be implemented to evaluate advertising copy, business documents, or literary works.

[0013] <<System configuration>> 1 is a schematic diagram showing a configuration of an information processing system according to the present embodiment. The information processing system according to the present embodiment includes an information processing device 1 and one or more user terminals 9, which are connected to a network and capable of communicating with each other. A user inputs a natural sentence to be analyzed to the information processing device 1 via the user terminal 9, and causes the information processing device 1 to analyze and evaluate the natural sentence. However, the system configuration for implementing the technology according to the present disclosure is not limited to the example in the present embodiment. The information processing system for implementing the technology according to the present disclosure may be implemented by a single information processing device.

[0014] The information processing device 1 is a computer including a central processing unit (CPU) 11, a read only memory (ROM) 12, a random access memory (RAM) 13, a storage device 14 such as an electrically erasable and programmable read only memory (EEPROM) or a hard disk drive (HDD), a communication unit 15 such as a network interface card (NIC), etc. However, the specific hardware configuration of the information processing device 1 can be omitted, replaced, or added as appropriate depending on the embodiment. In addition, the information processing device 1 is not limited to a device consisting of a single housing. The information processing device 1 may be realized by multiple devices using so-called cloud or distributed computing technology, etc.

[0015] The user terminal 9 is a terminal device used by a user. The user terminal 9 is a computer equipped with a CPU, a ROM, a RAM, a storage device, a communication unit, an input device, an output device, and the like (not shown). However, the specific hardware configuration of the user terminal 9 can be omitted, replaced, or added as appropriate depending on the embodiment. Furthermore, the user terminal 9 is not limited to a device consisting of a single housing. The user terminal 9 may be realized by a plurality of devices using so-called cloud or distributed computing technology, etc. The user uses the services provided by the system according to this embodiment via these user terminals 9.

[0016] 2 is a diagram showing an outline of the functional configuration of the information processing device 1 according to this embodiment. The information processing device 1 functions as an information processing device including a unit dividing section 21, an element identifying section 22, an element coordinate setting section 23a, a mass setting section 25, a parameter setting section 26, a displacement calculation section 27, and an evaluation value calculation section 28, by a program recorded in the storage device 14 being read into the RAM 13 and executed by the CPU 11, and each piece of hardware included in the information processing device 1 is controlled. Note that in this embodiment and other embodiments described later, each function included in the information processing device 1 is executed by the CPU 11, which is a general-purpose processor, but some or all of these functions may be executed by one or more dedicated processors.

[0017] The unit division unit 21 divides the natural text to be analyzed (hereinafter, referred to as "natural text to be analyzed") into a plurality of text units. As described above, in this embodiment, an example will be described in which a patent claim is the natural text to be analyzed. In recent years, the need for intellectual property value evaluation for business purposes (negotiations) and internal decision-making has increased with changes in the social and economic environment due to changes in business models and the increased role of intangible assets, and in order to accurately perform patent value evaluation, which is one type of intellectual property value evaluation, it is necessary to evaluate patent claims. However, a method capable of quantitatively evaluating the technical scope specified in a patent claim has not yet been established. For this reason, in this embodiment, a new patent modeling is adopted that uses "element decomposition of technical scope" by natural language processing based on AI / ML (machine learning) technology and "interaction evaluation between elements (compatibility between elements)" using Classical MD (molecular dynamics) calculation.

[0018] The element identification unit 22 performs a morpheme analysis of the natural sentence to be analyzed, extracts a plurality of morphemes constituting the natural sentence to be analyzed, and identifies the plurality of morphemes as a plurality of elements. However, in the technology according to the present disclosure, the objects treated as elements of natural sentences are not limited to morphemes. An element of natural sentences may be any part of natural sentences, and may be, for example, a paragraph, a clause, a phrase, a word, etc.

[0019] The element coordinate setting unit 23a sets virtual coordinates of a virtual particle when each of a plurality of elements constituting each of a plurality of sentence units is regarded as a virtual particle. In this embodiment, the element coordinate setting unit 23a inputs the element into a trained natural language model to obtain a distributed representation (vector representation) of the element. Then, the element coordinate setting unit 23a determines virtual coordinates of the virtual particle when each of the elements is regarded as a virtual particle by compressing the distributed representation of the element corresponding to the virtual particle to a predetermined dimension (e.g., three dimensions) by principal component analysis (PCA), and sets the determined virtual coordinates as the coordinates of the virtual particle.

[0020] The mass setting unit 25 regards the elements as a plurality of virtual particles placed at the virtual coordinates corresponding to each element, and sets a virtual mass to the plurality of virtual particles. In this embodiment, the mass setting unit 25 determines the abstraction level of the element corresponding to each virtual particle based on the distributed representation group of the elements corresponding to the plurality of virtual particles obtained by the element coordinate setting unit 23a, and sets a virtual mass according to the determined abstraction level.

[0021] The parameter setting unit 26 sets parameters for dynamics calculations according to environmental data related to the natural text to be analyzed. The type of environmental data to be referenced is not limited, but when the natural text to be analyzed is a patent claim, for example, data indicating the economic environment is adopted.

[0022] The displacement calculation unit 27 calculates the displacements of the multiple virtual particles by applying a potential function to the multiple virtual particles and performing dynamics calculations.

[0023] 3 is a diagram showing the concept of calculating the displacement after dynamics calculation by regarding elements constituting natural text as virtual particles in this embodiment. For example, if any constituent element (sentence unit) in a patent claim is a character string of "D which inputs A to B and determines C", the character string is decomposed into the elements of [A], [for], [B], [to], [input], [do], [C], [for], [determine], [do], and [D] by morphological analysis, and these elements are regarded as virtual particles, and dynamics calculation between particles is performed. In this manner, in this embodiment, the displacement after dynamics calculation is calculated for multiple elements included in the patent claim.

[0024] The evaluation value calculation unit 28 calculates an evaluation value indicating the interaction between the multiple elements that make up the natural text to be analyzed, based on the displacements calculated for the multiple virtual particles. Here, since the smaller the displacement of the virtual particles after the dynamics calculation, the higher the stability, it is preferable to adopt a scale for the evaluation value such that the smaller the displacement after the calculation, the higher the evaluation.

[0025] <<Process flow>> Next, a flow of processing executed by the information processing device according to the present embodiment will be described. Note that the specific contents and processing order of the processing described below are an example for implementing the present disclosure. The specific contents and processing order may be appropriately selected according to the embodiment of the present disclosure.

[0026] 4 is a flowchart showing the flow of the natural language analysis process according to the present embodiment. The process shown in this flowchart is executed when a command to start the natural language analysis process is issued after a natural language to be evaluated (natural language to be analyzed) is input by a user to the information processing device 1.

[0027] In step S101, the natural text to be analyzed is divided into one or more text units. The unit division unit 21 divides the natural text to be analyzed into a plurality of text units. In this embodiment, an example will be described in which a patent claim is the natural text to be analyzed, and so-called constituent features are analyzed as text units. Here, various units can be adopted as units into which the text units are divided according to the embodiment, and the units into which the text units are divided are not limited to the examples in this embodiment. For example, the text units may be divided based on some delimiter inserted in the text, such as a line break, a paragraph, or a punctuation mark, or may be divided by analyzing the structure or flow of the entire text. For example, in the case of a patent claim, the constituent features are often divided by line breaks, so it is easy to divide each constituent feature defined by the creator by line breaks or paragraphs into each text unit. However, since there are also patent claims that are not clearly divided, it is preferable that the method of dividing into text units is appropriately selected according to the structure of the natural text to be analyzed. Thereafter, the process proceeds to step S102.

[0028] In step S102, a morphological analysis of the natural sentence to be analyzed is performed, and a plurality of elements (components) are extracted from the sentence unit. In this embodiment, the morphemes constituting the natural sentence are used as the elements of the natural sentence to be analyzed, and the element identification unit 22 performs a morphological analysis of the natural sentence to be analyzed to extract a plurality of morphemes (tokens) constituting the natural sentence to be analyzed, and identifies the plurality of morphemes included in the sentence unit as a plurality of elements. For example, in the case of a natural sentence of "D which inputs A into B and determines C", the element identification unit 22 extracts morphemes such as [A], [for], [B], [to], [input], [do], [C], [for], [determine], [do], and [D] from the natural sentence, and sets these as a plurality of elements included in the natural sentence. Here, an existing Tokenizer (for example, MeCab, etc.) may be used for the morphological analysis. After that, the process proceeds to step S103.

[0029] In steps S103 and S104, virtual coordinates are assigned to each of the multiple elements when they are regarded as virtual particles. First, a distributed representation (vector representation) of each element is acquired via an encoder. The element coordinate setting unit 23a inputs each of the multiple elements constituting the target sentence unit to an encoder (e.g., a trained natural language processing model) to acquire an N-dimensional distributed representation output from the model (step S103). For example, "r 0[A] =[0.1, 0.2, 0.9, 0.7...]" is obtained. Here, an existing trained natural language processing model (e.g., Word2Vec, FastText, Transformer-Encoder, etc.) may be used as the natural language processing model. In addition, although an example in which a trained natural language processing model is used as the encoder has been described in this embodiment, other encoders may also be used.

[0030] When the distributed representation of each of the elements constituting the target sentence unit is acquired, virtual coordinates are assigned to each element based on the acquired distributed representation. The element coordinate setting unit 23a sets virtual coordinates of each element by compressing the distributed representation obtained in step S103 for each of the elements constituting the target sentence unit to a predetermined dimension (e.g., three dimensions) by principal component analysis (step S104). For example, 0[A] =[0.1, 0.2, 0.9, 0.7...]” is called r [A] =[0.2, 0.1, 0.4]". In principal component analysis, the dimension is lowered so that objects located close to each other in the distributed representation (N dimensions) before compression are also located close to each other in a lower predetermined dimension (here, three dimensions) that is the compression target. Note that, in this embodiment, an example of compression to three dimensions will be described, but the number of dimensions of the compression destination is not limited to three.

[0031] 5 is a schematic diagram showing a state in which a plurality of elements to which virtual coordinates are assigned are plotted in a three-dimensional space in this embodiment. In this embodiment, each element included in the natural text to be analyzed is regarded as a classical virtual particle (hard sphere). After the virtual coordinates are assigned to the virtual particle, the process proceeds to step S105.

[0032] In steps S105 and S106, a virtual mass is assigned to each of the multiple elements when it is regarded as a virtual particle. In this embodiment, the virtual mass of each element is determined based on the level of abstraction of each element determined based on the distributed representation group. The mass setting unit 25 determines the level of abstraction of the element corresponding to each virtual particle based on the distributed representation group of the multiple elements corresponding to the multiple virtual particles obtained in step S103 (step S105), and sets a virtual mass according to the determined level of abstraction (step S106).

[0033] 6 is a diagram showing a concept of setting a virtual mass to a virtual particle corresponding to an element included in a natural text to be analyzed in this embodiment. First, the mass setting unit 25 classifies elements extracted from the natural text to be analyzed into categories based on the distributed representation of each element. Here, the category classification of elements may use a method of classifying elements into categories set in advance in the dictionary by comparing them with data in a dictionary, or may use a method of clustering similar elements using machine learning.

[0034] When the elements are classified into a plurality of categories, the center of gravity of the category is set for each category. The center of gravity of the category may be calculated based on virtual coordinates given to the elements included in the target category, or may be predefined for each category. When the center of gravity of the category is set, the distance from the center of gravity (Euclidean distance) is calculated for each virtual particle corresponding to the elements included in the category. Then, the mass setting unit 25 sets a virtual mass according to the distance from the center of gravity for each virtual particle. The virtual mass may be calculated by inputting the distance into a function having a predetermined coefficient, and the coefficient and function may be such that the virtual mass becomes smaller the farther from the center of gravity, and the virtual mass becomes larger the closer to the center of gravity. That is, the virtual mass is set so that the mass is attenuated according to the Euclidean distance from the center of gravity. This is because it is estimated that the element related to the virtual particle farther from the center of gravity is lower in abstraction (higher in concreteness), and the element related to the virtual particle closer to the center of gravity is higher in abstraction (lower in concreteness). When the virtual mass is set for the virtual particle, the process proceeds to step S107.

[0035] In step S107, parameters for the dynamics calculation are set. The parameter setting unit 26 sets parameters for the dynamics calculation according to environmental data related to the natural text to be analyzed. In this embodiment, since an example is described in which a patent claim is the natural text to be analyzed, the parameter setting unit 26 sets calculation parameters (such as temperature) in the dynamics calculation according to economic conditions such as the market size and market share of the business domain corresponding to the patent claim. When setting calculation parameters for temperature based on the market share, for example, a temperature setting may be performed in which a market share of 100% is set to 0K (absolute zero) and the temperature is increased according to a reduction in the share. This is because the lower the temperature condition of the simulation, the less likely the particles are to be displaced, and the higher the temperature condition of the simulation, the more likely the particles are to be displaced. What temperature is set for a market share of 0% and how is the function for increasing the temperature according to a reduction in the share (which may be linear or may be an exponential function) set can be arbitrarily set according to the embodiment. When determining parameters based on environmental data other than the market share, the parameters can be set with a similar idea. That is, the parameters can be obtained by inputting the environmental data into a predetermined function that expresses the relationship between the environment related to the natural text to be analyzed and the ease of particle displacement as the relationship between the environmental data and the parameters. In the present embodiment, an example in which temperature is set as a parameter has been described, but other parameters such as pressure may also be set according to the environmental data related to the natural text to be analyzed. After that, the process proceeds to step S108.

[0036] In step S108, for each of the multiple sentence units, dynamics calculation is performed when each of the multiple elements constituting the sentence unit is regarded as a virtual particle. The displacement calculation unit 27 regards the multiple elements constituting the target sentence unit as multiple virtual particles having virtual masses placed at virtual coordinates corresponding to each element, and calculates the displacement of the multiple virtual particles corresponding to the multiple elements by applying a potential function to the multiple virtual particles and performing dynamics calculation. More specifically, each of the multiple elements constituting the target sentence unit is regarded as a virtual particle located at a virtual coordinate, and the displacement amount of each virtual particle after a predetermined time has elapsed when an arbitrary potential function is applied with the number of particles, volume, and temperature constant (canonical (NVT) ensemble) is calculated, thereby calculating the displacement between elements in the sentence unit after the dynamics calculation. Here, the predetermined time may be set to any time between several picoseconds and 1 nanosecond, for example.

[0037] The formula shown below is a formula showing a molecular dynamics (MD) calculation, and shows that the internal energy of the system shown on the left side (corresponding to the Hamiltonian of the system) is the sum of the kinetic energy of the system shown on the right side (energy corresponding to the momentum p of each virtual particle) and the potential energy of the system (interaction energy corresponding to the position r of each virtual particle). In the technology shown in this embodiment, each of the multiple elements that make up a sentence unit is regarded as a virtual particle, and one sentence unit is regarded as one system to perform dynamics calculations.

number

[0038] FIG. 7 is a schematic diagram showing a state in which a dynamics calculation is performed by applying a potential function between virtual particles corresponding to a plurality of elements constituting a natural sentence in this embodiment. In this embodiment, a virtual spring is set between the virtual particles, and the system is treated as a harmonic oscillator. Here, there is a most stable bond length between two particles (e.g., Lennard-Jones Potential), and there is a most stable bond angle between three particles (e.g., Stillinger-Weber Potential). In the dynamics calculation, the displacement of the virtual particle under the set conditions (various parameters including virtual coordinates, virtual mass, and temperature) is calculated using a potential function. A method for determining the potential function to be applied between virtual particles will be described later. When the displacement after the dynamics calculation for each of the plurality of elements is calculated, the process proceeds to step S109.

[0039] In steps S109 and S110, the stability of the sentence units and the evaluation value of the natural sentence to be analyzed are calculated. The evaluation value calculation unit 28 calculates an evaluation value indicating the interaction between the multiple elements constituting each of the multiple sentence units based on the displacements calculated for the multiple virtual particles included in the sentence unit (step S109). Then, the evaluation value calculation unit 28 calculates an evaluation value of the entire natural sentence to be analyzed based on the stability of each sentence unit (step S110). Specifically, the sum, simple average, weighted average, etc. of the evaluation values ​​calculated for the multiple sentence units may be adopted as the evaluation value of the entire natural sentence to be analyzed. When the weighted average is adopted, it is necessary to set a weight for each sentence unit, and the weight of the sentence unit may be determined based on the length of the sentence unit, the sum of the virtual masses of the elements included in the sentence unit, etc.

[0040] FIG. 8 is a diagram showing a concept of performing an evaluation according to the amount of displacement in this embodiment. Since the smaller the displacement of the virtual particle after the dynamics calculation, the more stable the particle, the smaller the displacement, the higher the evaluation. For example, the exponential of the total amount of displacement may be used as the evaluation value. In this case, for example, the evaluation may be performed such that the maximum evaluation value is set when the displacement is 0 (zero), and the evaluation value decreases as the displacement increases. This is because the higher the structural stability, the smaller the amount of displacement of the particle, and the lower the structural stability, the larger the amount of displacement of the particle. As the amount of displacement referred to for the evaluation, it is possible to refer to the distance between the virtual particles. In addition, as the amount of displacement referred to for the evaluation, the amount of change in the angle formed by the three virtual particles may also be referred to.

[0041] However, the specific expression method and scale of the evaluation value are not limited to the examples in this disclosure. The evaluation value may be calculated according to the total amount of displacement, and may be a number, a rank, an estimated patent transfer price, or the like. When the evaluation value of the entire natural text to be analyzed is calculated, the process shown in this flowchart ends.

[0042] Next, a method for determining a potential function to be applied between virtual particles will be described. Although the method for determining the potential function is not limited to the specific example described in this embodiment, it is preferable that the potential function and parameters to be applied are determined according to a combination of at least any of the attributes of the natural text to be analyzed that corresponds to the virtual particle to which the potential function is to be applied, the corresponding element, and the corresponding sentence unit.

[0043] In this embodiment, an example will be described in which a patent classification code (e.g., the International Patent Classification (IPC) or FI used by the Japan Patent Office, etc.) is used as an attribute of an element or sentence unit corresponding to each of a plurality of virtual particles to which a potential function is applied. The method of identifying the patent classification code corresponding to an element or sentence unit is not limited, but in addition to a method of searching a list of patent classification codes, a method of inputting an element or sentence unit into a classifier (classification model) generated by subjecting a relationship between a word, sentence, etc. and a patent classification code to supervised machine learning, and obtaining a patent classification code as an output, may be adopted.

[0044] FIG. 9 is a diagram showing an overview of a process for acquiring attributes of elements of natural language using a machine learning model in this embodiment. In the example shown in FIG. 9, the word "payment" as element E1 and the word "neural network" as element E2 are input to a multi-label classifier, and one or more labels predicted by the classifier and the probability of each label are obtained as the output. In the example shown in FIG. 9, [···G06Q10 / 00, G06Q20 / 00, G06Q30 / 00···] are obtained as patent classification codes (labels) of element E1 "payment", and [···0.0, 0.7, 0.3···] are obtained as the probabilities of these patent classification codes. In addition, [···G06N3 / 02···] is obtained as the patent classification code (label) of element E2 "neural network", and [···0.9···] is obtained as the probability of these patent classification codes. When the prediction result is obtained by the classifier, the displacement calculation unit 27 reads out the potential function and parameters preset for the combination of the obtained patent classification codes, and applies them between the target virtual particles. Note that, although Fig. 9 illustrates an example in which there are two virtual particles (2 bodies), the process flow is the same even when there are three or more virtual particles (3 bodies).

[0045] As the potential function, for example, the Lennard-Jones Potential function shown below for calculating an interatomic potential can be adopted. Here, the parameter ε indicates the depth of the potential well, and the parameter σ indicates the distance at which the potential value becomes zero. Note that, although an example in which a relatively simple function such as the Lennard-Jones Potential is adopted is described in this embodiment, the functions that can be adopted are not limited to the example below.

number

[0046] Next, a method for determining the parameters ε and σ when the above-mentioned function is used to calculate the interatomic potential will be described with reference to the drawings.

[0047] FIG. 10 is a diagram showing an outline of the relationship between the distance r between virtual particles and the energy V(r) when a function for calculating an interatomic potential is adopted as the potential function in this embodiment. In addition, in this figure, a graph is shown in which the exponent part (12,6) in the function is changed to (4,2) for ease of understanding. The dashed line C1 shows the relationship between the distance r between these virtual particles P1 and P2 and the energy V(r) when the virtual particle P1 (see FIG. 9) corresponding to the element E1 (in the example shown in FIG. 9, "Settlement") has the attribute A1 (in the example shown in FIG. 9, patent classification code G06Q20 / 00), and the virtual particle P2 (see FIG. 9) corresponding to the element E2 (in the example shown in FIG. 9, "Neural Network") has the attribute A2 (in the example shown in FIG. 9, patent classification code G06N3 / 02). The potential function used here is a function for calculating the interatomic potential as described above, and since the interatomic potential is determined by the energy of repulsive interaction and the energy of attractive interaction, both of which change according to distance, the point where the energy V(r) is a minimum value appears in the middle of the graph. And, since the lower the energy, the more stable it is, the distance r where the energy V(r) is a minimum value is the distance where the evaluation is highest in the relationship between element E1 and element E2.

[0048] Fig. 11 is a diagram showing an example of parameters set in the potential function when a function for calculating an interatomic potential is adopted as the potential function in this embodiment. The displacement calculation unit 27 determines the parameters to be applied to the potential function by referring to parameters defined in advance for a combination of attributes (attribute pair). When the referenced patent classification code is the most probable patent classification code (i.e., a representative label) for each virtual particle, for example, a combination of attribute A1 (patent classification code G06Q20 / 00) and attribute A2 (patent classification code G06N3 / 02), according to the example shown in Fig. 11, the displacement calculation unit 27 reads out parameters ε = 0.5 and σ = 1.0 as parameters defined in advance for the combination of these attributes, and applies these parameters to the potential function to perform dynamics calculation.

[0049] The patent classification code referenced at this time may be only the patent classification code with the highest probability for each virtual particle (i.e., a representative label), or in addition to the representative patent classification code, multiple patent classification codes including patent classification codes with probabilities ranked second or lower may be referenced. The method of dynamics calculation when patent classification codes ranked second or lower in probability are also referenced is not limited to the example in this embodiment, but below, a method of using a simple average or weighted average of parameters of a potential function when referring to patent classification codes ranked second or lower will be described.

[0050] Here, an example of a method for determining parameters of a potential function when dynamics calculation is performed for a virtual particle P1 corresponding to element E1 "payment" taking into consideration not only the attribute A1 (patent classification code G06Q20 / 00) with the highest probability, but also the attribute A3 (patent classification code G06Q30 / 00) with the second highest probability. In Fig. 10, the dashed line C2 indicates the relationship between the distance r and the energy V(r) between the virtual particles P1 and P2 when the virtual particle P1 corresponding to element E1 "payment" has the attribute A3 (patent classification code G06Q30 / 00 in the example shown in Fig. 9) and the virtual particle P2 corresponding to element E2 "neural network" has the attribute A2 (patent classification code G06N3 / 02).

[0051] Here, if only the attribute A3 of the attributes of the virtual particle P1 is taken into consideration, the parameter ε=0.9 and the parameter σ=1.2 are read out as parameters defined in advance for the combination of these attributes (see FIG. 11), and these parameters are applied to the potential function to perform the dynamics calculation, but here, the attributes A1 and A3 of the attributes of the virtual particle P1 are taken into consideration. For this reason, for the parameter ε and the parameter σ, values ​​calculated based on both the parameter ε=0.5 and the parameter σ=1.0 when the attribute of the virtual particle P1 is the attribute A1, and the parameter ε=0.9 and the parameter σ=1.2 when the attribute of the virtual particle P1 is the attribute A3 are adopted.

[0052] When a simple average is adopted as the method for determining the parameters, the parameters applied to the potential function between virtual particle P1 and virtual particle P2 are the simple averages of parameters ε=0.7 and σ=1.1 (see FIG. 11). When such simple average parameters are adopted, the curve showing the relationship between the distance r between the virtual particles and the energy V(r) is intermediate between the curve for the first attribute pair (in the example shown in the figure, the pair of attributes A1 and A2) and the curve for the second attribute pair (in the example shown in the figure, the pair of attributes A3 and A2) (see solid line C3 in FIG. 10).

[0053] On the other hand, a weighted average may be adopted as a method for determining the parameters. Various values ​​can be adopted as the weights when calculating the weighted average, but here, an example will be described in which the weighted average is calculated based on the probability output from the multi-label classifier for the target label. Specifically, the displacement calculation unit 27 calculates the parameters ε and σ using the following equations so that the curve showing the relationship between the distance r between the virtual particles and the energy V(r) approaches the curve with the larger weight between the curve for the first attribute pair and the curve for the second attribute pair. Weighted average parameter = (first attribute pair parameter - second attribute pair parameter) * weighting factor + second attribute pair parameter Weighting factor = probability of first attribute pair / (probability of first attribute pair + probability of second attribute pair)

[0054] Applying the example parameters shown in Figure 11 to the above formula, we get Parameter ε=(0.5-0.9)*0.7+0.9 Parameter σ=(1.0-1.2)*0.7+1.2 and parameters ε=0.62 and σ=1.06, which are weighted averages, are adopted (see FIG. 11). The curve showing the relationship between the distance r between virtual particles and the energy V(r) is biased toward the curve with the greater weight between the curve for the first attribute pair and the curve for the second attribute pair (see solid line C4 in FIG. 10).

[0055] That is, in this embodiment, when two attributes G06Q20 / 00 and G06Q30 / 00 are estimated for an element constituting a natural sentence "payment", the final parameters between "payment" and "neural network" are set for the virtual particle pair "payment"-"neural network" by considering not only the parameters of attribute A1 (G06Q20 / 00)-attribute A2 (G06N3 / 2) but also the parameters of attribute A3 (G06Q30 / 00)-attribute A2 (G06N3 / 2). Here, an example in which two or more attributes are considered for one of the two virtual particles has been described, but the basic calculation method is the same when two or more attributes are considered for both of the two virtual particles. The basic calculation method is also the same when three or more attributes are considered for the virtual particles, or when multiple attributes are considered for three or more virtual particles. Here, as for the weight, only the probability (estimated result) of the attribute of one virtual particle may be used, or the product of the probabilities of the attributes of both virtual particles may be used as the weight.

[0056] According to the system of the present embodiment described above, it is possible to quantitatively evaluate natural language (in the example shown in the present embodiment, a patent claim). More specifically, according to the system of the present embodiment described above, it is possible to evaluate the stability of natural language by evaluating the interaction / compatibility between multiple elements that make up the natural language.

[0057] Second Embodiment In the first embodiment described above, an example in which the technology according to the present disclosure is used to evaluate patent claims has been described, but the technology according to the present disclosure can be used to evaluate any natural language, and the scope of application is not limited to the evaluation of patent claims. For example, the technology according to the present disclosure can be used to evaluate advertising copy. Note that the hardware configuration of the system according to this embodiment is roughly the same as that of the first embodiment, and therefore will not be described.

[0058] 12 is a diagram showing an outline of the functional configuration of the information processing device 1b according to this embodiment. The information processing device 1b functions as an information processing device including an element identification unit 22, an element coordinate setting unit 23a, a mass setting unit 25, a parameter setting unit 26, a displacement calculation unit 27, and an evaluation value calculation unit 28, by a program recorded in the storage device 14 being read into the RAM 13 and executed by the CPU 11, and each piece of hardware included in the information processing device 1b is controlled. Note that in this embodiment and other embodiments described later, each function included in the information processing device 1b is executed by the CPU 11, which is a general-purpose processor, but some or all of these functions may be executed by one or more dedicated processors.

[0059] The information processing device 1b according to this embodiment differs from the information processing device 1 according to the first embodiment in that the unit dividing unit 21 in the first embodiment is omitted, and that the displacement calculation unit 27 and the evaluation value calculation unit 28 calculate the displacement and evaluation value of the entire natural sentence to be analyzed without calculating the displacement and evaluation value for each sentence unit. However, for natural sentences other than patent claims, such as advertising copy, the energy may be calculated by dividing the sentences into multiple stages, such as sentence units and elements, as in the first embodiment, and the evaluation value may be obtained.

[0060] 13 is a flowchart showing the flow of the natural language analysis process according to this embodiment. The process shown in this flowchart is executed when a user inputs a natural language sentence to be evaluated (a natural language sentence to be analyzed) into the information processing device 1b and then issues an instruction to start the natural language analysis process.

[0061] In step S201, a morphological analysis of the natural sentence to be analyzed is performed, and a plurality of elements (components) are extracted from the natural sentence to be analyzed. In this embodiment, as in the first embodiment described above, the morphemes constituting the natural sentence are used as elements of the natural sentence to be analyzed, and the element identification unit 22 performs a morphological analysis of the natural sentence to be analyzed to extract a plurality of morphemes (tokens) constituting the natural sentence to be analyzed, and identifies the plurality of morphemes included in the natural sentence to be analyzed as a plurality of elements. For example, in the case of a natural sentence "You can get a lot of points," the element identification unit 22 extracts morphemes such as [points], [a lot], and [get] from the natural sentence, and sets these as a plurality of elements included in the natural sentence. After that, the process proceeds to step S202.

[0062] In steps S202 and S203, virtual coordinates are assigned to each of the elements when the elements are regarded as virtual particles. In this embodiment, as in the first embodiment described above, the element coordinate setting unit 23a inputs each of the elements constituting the natural text to be analyzed to an encoder to obtain an N-dimensional distributed representation output from the model (step S202). Then, the element coordinate setting unit 23a sets virtual coordinates of each of the elements constituting the natural text to be analyzed by compressing the distributed representation obtained in step S202 to a predetermined dimension by principal component analysis (step S203). When the virtual coordinates are assigned to the virtual particles, the process proceeds to step S204.

[0063] In steps S204 and S205, a virtual mass is assigned to each of the elements when the elements are regarded as virtual particles. In this embodiment, the virtual mass of each element is determined based on the level of abstraction of each element determined based on the distributed representation group. The mass setting unit 25 determines the level of abstraction of the element corresponding to each virtual particle based on the distributed representation group of the elements corresponding to the virtual particles obtained in step S103 (step S204), and sets a virtual mass according to the determined level of abstraction (step S205). A more specific method of assigning a virtual mass to a virtual particle can be substantially the same as that described in the first embodiment, and therefore a description thereof will be omitted. Thereafter, the process proceeds to step S206.

[0064] In step S206, parameters for dynamics calculation are set according to environmental data related to the natural text to be analyzed. The parameter setting unit 26 sets parameters for dynamics calculation according to environmental data related to the natural text to be analyzed. In this embodiment, an example is described in which advertising copy is the natural text to be analyzed, so the parameter setting unit 26 sets calculation parameters (temperature, etc.) in dynamics calculation according to environmental data of the business domain corresponding to the advertising copy. In this embodiment, an example is described in which temperature is set as a parameter, but other parameters such as pressure may also be set according to environmental data related to the natural text to be analyzed. Thereafter, the process proceeds to step S207.

[0065] In step S207, dynamics calculation is performed when each of the elements constituting the natural text to be analyzed is regarded as a virtual particle. The displacement calculation unit 27 regards the elements constituting the natural text to be analyzed as a virtual particle having a virtual mass placed at virtual coordinates corresponding to each element, and calculates the displacement of the virtual particles corresponding to the elements by applying a potential function to the virtual particles. A more specific method of dynamics calculation can be roughly the same as that described in the first embodiment, so the description will be omitted. After that, the process proceeds to step S208.

[0066] In step S208, the stability of the natural text to be analyzed and an evaluation value of the natural text to be analyzed are calculated. The evaluation value calculation unit 28 calculates an evaluation value indicating the stability of the natural text to be analyzed based on the displacements calculated for the multiple virtual particles included in the natural text to be analyzed. A more specific method for calculating the evaluation value can be substantially the same as that described in the first embodiment, and therefore a description thereof will be omitted. After that, the process shown in this flowchart ends.

[0067] According to the system of the present embodiment described above, it is possible to quantitatively evaluate natural language (advertising copy in the example shown in the present embodiment). More specifically, according to the system of the present embodiment described above, it is possible to evaluate the stability of natural language by evaluating the interaction / compatibility between multiple elements that make up the natural language.

[0068] <Third embodiment> In the first embodiment described above, an example was described in which virtual mass was assigned to virtual particles corresponding to an element according to the abstraction level of the element, but virtual mass may be assigned to a sentence unit, which is a larger group than an element. In this embodiment, a potential function is applied to virtual particles of multiple elements included in the sentence unit to calculate the potential energy of the sentence unit, and a virtual mass according to the calculated potential energy is assigned to the sentence unit. Note that the hardware configuration of the system according to this embodiment is roughly the same as that of the first embodiment, so a description thereof will be omitted.

[0069] 14 is a diagram showing an outline of the functional configuration of the information processing device 1c according to this embodiment. The information processing device 1c functions as an information processing device including a unit dividing unit 21, an element identifying unit 22, an element coordinate setting unit 23a, a unit coordinate setting unit 23b, an inter-element energy calculation unit 24a, a mass setting unit 25, a parameter setting unit 26, a displacement calculation unit 27, and an evaluation value calculation unit 28, by reading a program recorded in the storage device 14 into the RAM 13 and executing it by the CPU 11, and controlling each piece of hardware included in the information processing device 1c. Note that in this embodiment and other embodiments described later, each function included in the information processing device 1c is executed by the CPU 11, which is a general-purpose processor, but some or all of these functions may be executed by one or more dedicated processors.

[0070] The information processing device 1c of this embodiment differs from the information processing device 1 of the first embodiment in that it further includes a unit coordinate setting unit 23b and an inter-element energy calculation unit 24a, and in that the processing targets of the mass setting unit 25 and the displacement calculation unit 27 are sentence units rather than elements.

[0071] The unit coordinate setting section 23b sets, for each of a plurality of sentence units constituting the natural sentence to be analyzed, virtual coordinates of the sentence unit based on virtual coordinates of a plurality of elements constituting the sentence unit.

[0072] The inter-element energy calculation unit 24a regards, for each of the multiple sentence units, the multiple elements that make up the target sentence unit as multiple virtual particles placed at virtual coordinates corresponding to each element, and calculates potential energy (hereinafter simply referred to as "energy") that indicates the interaction between the virtual particles corresponding to the multiple elements by applying a potential function to the multiple virtual particles and performing a force field calculation.

[0073] 15 is a diagram showing the concept of calculating energy by regarding elements constituting natural text as virtual particles in this embodiment. For example, if one of the constituent features (text units) in a patent claim is a character string "D, which inputs A into B and determines C", the character string is decomposed into the elements [A], [for], [B], [to], [input], [do], [C], [for], [determine], [do], and [D] by morphological analysis, and these elements are regarded as virtual particles, and force field calculation between particles is performed. In this manner, in this embodiment, the energy for each of a plurality of text units is calculated.

[0074] Here, the energy calculation performed by the inter-element energy calculation unit 24a to assign virtual mass to a sentence unit differs from a molecular dynamics calculation in that it is a static force field calculation in which kinetic energy is calculated as 0 (ignored) and the potential energy when one sentence unit is considered as a system is treated as the internal energy of the system.

[0075] The mass setting unit 25 regards the multiple sentence units as multiple virtual particles placed at virtual coordinates corresponding to each sentence unit, and sets a virtual mass for each of the multiple sentence units based on the energy indicating the interaction between the multiple virtual particles corresponding to the multiple elements that make up the target sentence unit.

[0076] The displacement calculation unit 27 calculates the displacements of the virtual particles corresponding to the sentence units by applying a potential function to the virtual particles corresponding to the sentence units and performing dynamics calculations. The dynamics calculations performed here are normal dynamics calculations in which the kinetic energy is not set to 0 (zero), unlike the force field calculations performed by the inter-element energy calculation unit 24a.

[0077] 16 is a flowchart showing the flow of the natural language analysis process according to this embodiment. The process shown in this flowchart is executed when a user inputs a natural language sentence to be evaluated (a natural language sentence to be analyzed) into the information processing device 1c and then instructs the start of the natural language analysis process.

[0078] The process from step S301 to step S304 is similar to the process from step S101 to step S104 described with reference to Fig. 4, and therefore the description will be omitted. After that, the process proceeds to step S305.

[0079] In step S305, virtual coordinates are assigned to each of the sentence units constituting the natural sentence to be analyzed, assuming that each of the sentence units is a virtual particle. Here, virtual coordinates are assigned to each sentence unit according to the virtual coordinates of each element included in the sentence unit.

[0080] FIG. 17 is a schematic diagram showing how a sentence unit including a plurality of elements is regarded as a virtual particle including virtual particles corresponding to the plurality of sentence units in this embodiment. The unit coordinate setting unit 23b sets virtual coordinates of each of the plurality of sentence units constituting the natural sentence to be analyzed, based on the virtual coordinates of the plurality of elements constituting the sentence unit. Various methods may be used to calculate the virtual coordinates assigned to the virtual particle corresponding to the sentence unit, but in this embodiment, the target sentence unit is coarse-grained, and a classical virtual particle (hard sphere) including all the elements included in the sentence unit is regarded as the virtual particle corresponding to the sentence unit. The virtual coordinates of the virtual particle corresponding to the sentence unit may be the coordinates of the center of a sphere based on the average of the virtual coordinates corresponding to the virtual particles of the plurality of elements, or may be the virtual coordinates of the center of gravity calculated in consideration of the coordinates of the virtual particles corresponding to the plurality of elements contained in the sphere and weights according to the potential energy generated for each of the virtual particles corresponding to the plurality of elements. Here, for example, the sum of potential energies calculated for all particle pairs including a virtual particle corresponding to a certain element may be adopted as the "potential energy generated for the virtual particle" corresponding to the element, and weighting may be performed according to the ranking of this sum. More specifically, if there are virtual particles A, B, and C corresponding to an element, and the potential energies are highest in the order of virtual particle pair AB, virtual particle pair BC, and virtual particle pair AC, the weight of virtual particle B, which has the largest sum of potential energies calculated for all particle pairs including itself, may be set to the largest value. After that, the process proceeds to step S306.

[0081] In step S306, for each of the multiple sentence units, energy indicating the interaction between the virtual particles when each of the multiple elements constituting the sentence unit is regarded as a virtual particle is calculated. Here, the potential energy between the elements in the sentence unit is calculated by applying an arbitrary potential function to the virtual coordinates of each element. The inter-element energy calculation unit 24a regards the multiple elements constituting the target sentence unit as multiple virtual particles placed at the virtual coordinates corresponding to each element, and calculates the energy indicating the interaction between the virtual particles corresponding to the multiple elements by applying the potential function to the multiple virtual particles and performing a force field calculation.

[0082] The equations used in the force field calculation here may be the same as those representing the molecular dynamics (MD) calculation described above. However, the energy calculation performed here differs from the molecular dynamics calculation described above in that it is a static force field calculation in which kinetic energy is calculated (ignored) as 0 (zero) and the potential energy when one sentence unit is regarded as one system is treated as the internal energy of the system. For this reason, the calculation performed in this step by applying a potential function to multiple virtual particles is referred to as a "force field calculation." After that, the process proceeds to step S307.

[0083] In step S307, a virtual mass based on potential energy is assigned to each of the multiple sentence units. As described above in step S305, each of the multiple sentence units is coarse-grained and considered as a classical multiple virtual particles (hard spheres). Here, the mass setting unit 25 sets a virtual mass to the virtual particle corresponding to each of the multiple sentence units based on the potential energy of the system corresponding to the sentence unit calculated in step S306. The virtual mass may be calculated by inputting the potential energy into a function having a predetermined coefficient, and the coefficient and function are used such that the lower the potential energy is, the larger the virtual mass is, and the higher the potential energy is, the smaller the virtual mass is. That is, the virtual mass is set so that the mass attenuates according to the magnitude of the energy. This is because it is estimated that the virtual particle with lower internal energy is more stable and therefore has a larger virtual mass, and the virtual particle with higher internal energy is more unstable and therefore has a smaller virtual mass. In addition, the function for calculating the virtual mass may be linear or exponential in the relationship between energy and mass. Thereafter, the process proceeds to step S308.

[0084] In step S308, parameters for dynamics calculation are set according to the environmental data related to the natural text to be analyzed. A more specific method for parameter setting can be roughly the same as that described in the first embodiment, so a description thereof will be omitted. Then, the process proceeds to step S309.

[0085] In step S309, dynamics calculation is performed when each of the multiple sentence units constituting the natural sentence to be analyzed is regarded as a virtual particle. The displacement calculation unit 27 regards the multiple sentence units constituting the natural sentence to be analyzed as multiple virtual particles having virtual masses placed at virtual coordinates corresponding to each sentence unit, and calculates the displacement of the multiple virtual particles corresponding to the multiple sentence units by applying a potential function to the multiple virtual particles and performing dynamics calculation. A more specific method of dynamics calculation can be roughly the same as that described in the first embodiment, so the description will be omitted. After that, the process proceeds to step S310.

[0086] In step S310, the stability of the natural text to be analyzed and an evaluation value of the natural text to be analyzed are calculated. The evaluation value calculation unit 28 calculates an evaluation value indicating the stability of the natural text to be analyzed based on the displacements calculated for the multiple virtual particles included in the natural text to be analyzed. A more specific method for calculating the evaluation value can be substantially the same as that described in the first embodiment, and therefore a description thereof will be omitted. After that, the process shown in this flowchart ends.

[0087] <Fourth embodiment> According to the first to third embodiments described above, it is possible to obtain an evaluation value of the natural text to be analyzed, but it is also possible to improve the evaluation value of the natural text to be analyzed by searching for replacement candidates for elements related to displaced virtual particles and replacing the elements as necessary. In this embodiment, the contents of the process when evaluating patent claims and searching for replacement candidates from the specification will be mainly described. Note that the hardware configuration of the system according to this embodiment is roughly the same as that of the first embodiment, so a description thereof will be omitted.

[0088] 18 is a diagram showing an outline of the functional configuration of an information processing device 1d according to this embodiment. The information processing device 1d functions as an information processing device including a unit dividing section 21, an element identifying section 22, an element coordinate setting section 23a, a mass setting section 25, a parameter setting section 26, a displacement calculating section 27, a fixing section 29, a searching section 30, and a correcting section 31, by reading a program recorded in the storage device 14 into the RAM 13 and executing it by the CPU 11, and controlling each piece of hardware included in the information processing device 1d. Note that in this embodiment and other embodiments described later, each function included in the information processing device 1d is executed by the CPU 11, which is a general-purpose processor, but some or all of these functions may be executed by one or more dedicated processors.

[0089] The information processing device 1d of this embodiment differs from the information processing device 1 of the first embodiment in that the evaluation value calculation unit 28 in the first embodiment can be omitted (the evaluation value calculation unit 28 may be provided), and in that it further includes a fixing unit 29, a search unit 30 and a correction unit 31.

[0090] The fixing unit 29 fixes at least one of the plurality of virtual particles, thereby preventing the fixed virtual particle from being displaced in the dynamics calculation by the displacement calculation unit 27. More specifically, in this embodiment, the fixing unit 29 identifies an element that is not a case component (a noun phrase required by a predicate in a sentence; a token) by analyzing a case dominance relationship of each element, and fixes at least a virtual particle corresponding to an element that is not a case component among the plurality of virtual particles, thereby preventing the virtual particle fixed in the dynamics calculation by the displacement calculation unit 27 from being displaced. However, the fixing unit 29 only needs to identify an element to be fixed by analyzing a relationship between elements (e.g., a subject, an object, etc.), and the analysis method for identifying an element to be fixed is not limited to the case dominance relationship.

[0091] The search unit 30 searches for an element that is similar to or coincides with the virtual coordinates after the displacement of the virtual particle by searching a group of elements (eg, a description) different from the natural text to be analyzed.

[0092] The correcting unit 31 corrects the element in the natural sentence to be analyzed that corresponds to the displaced virtual particle by using the element retrieved by the searching unit 30 .

[0093] 19 is a flowchart showing the flow of the natural language analysis process according to this embodiment. The process shown in this flowchart is executed when a user inputs a natural language sentence to be evaluated (a natural language sentence to be analyzed) into the information processing device 1d and then instructs the start of the natural language analysis process.

[0094] The processing from step S401 to step S407 is similar to the processing from step S101 to step S107 described with reference to Fig. 4, and therefore the description will be omitted. After that, the processing proceeds to step S408.

[0095] In step S408, virtual particles corresponding to some of the elements are fixed so as not to be displaced in accordance with each dominance relationship between the elements. The fixing unit 29 fixes at least one of the virtual particles so that the fixed virtual particle is not displaced by the dynamics calculation by the displacement calculation unit 27.

[0096] FIG. 20 is a diagram showing an example of the analysis of the case dominance relationship in this embodiment. As described above, in this embodiment, the fixing unit 29 identifies elements that are not case components by analyzing the case dominance relationship of each element, and fixes at least virtual particles corresponding to elements that are not case components among the multiple virtual particles, thereby preventing the virtual particles corresponding to elements that are not case components from being displaced by the dynamics calculation by the displacement calculation unit 27. According to the example shown in the figure, if the natural sentence to be analyzed has a character string "D, which inputs A into B and determines C", this is classified as follows: case component (target word) "[A], [wo]", case component (positional case) "[B], [ni]", action "[input], [shi]", case component (target word) "[C], [wo]", action "[determine], [do]", and constituent "[D]". Here, a trained large-scale natural language model may be used for the analysis of the classification task. Then, the fixing unit 29 fixes the virtual particles corresponding to the components other than the case components so as not to displace them during the dynamics calculation, and displaces the virtual particles corresponding to at least some of the case components during the dynamics calculation. When the virtual particles to be fixed are determined, the process proceeds to step S409.

[0097] In step S409, for each of the multiple sentence units, dynamics calculation is performed in the case where each of the multiple elements constituting the sentence unit is regarded as a virtual particle. The displacement calculation unit 27 regards the multiple elements constituting the target sentence unit as multiple virtual particles having virtual masses placed at virtual coordinates corresponding to each element, and calculates the displacement of the multiple virtual particles corresponding to the multiple elements by applying a potential function to the multiple virtual particles. A more specific method of dynamics calculation can be roughly the same as the method described in the first embodiment, except that calculation is performed so that the fixed virtual particles are not displaced, so a description will be omitted.

[0098] 21 is a diagram showing the concept of dynamics calculation in a state where some virtual particles are fixed in this embodiment. In this embodiment, dynamics calculation is performed so that the virtual particles set as fixed targets in step S408 are not displaced. Then, the process proceeds to step S410.

[0099] In step S410, replacement candidates for the element corresponding to the displaced element are searched from other documents, and the displaced element is replaced with the replacement candidates. The search unit 30 searches for elements that are similar to or match the virtual coordinates after the displacement of the virtual particle by searching a group of elements different from the natural text to be analyzed. In this embodiment, an example is described in which the natural text to be analyzed is a patent claim, so that the group of elements to be searched for can be, for example, a patent specification related to the application of the patent claim. In this way, the search unit 30 can search for replacement candidates for the element of the patent claim within a range where the problem of so-called support requirements does not occur. However, the group of elements to be searched for may be any group of elements that includes elements that can replace the element of the natural text to be analyzed, and may be, for example, any document, dictionary, database, the Internet, etc.

[0100] A distributed representation of an element may be used for the search. That is, the search unit 30 acquires in advance a distributed representation corresponding to a phrase (element) in an element group (e.g., a specification) to be searched for, and a vector (virtual coordinate) after dimensional compression of the distributed representation. Then, the search unit 30 searches for an element in the element group that is closest to the vector (virtual coordinate) after displacement as a replacement candidate by matching the vector (virtual coordinate) of the element in the element group with the vector (virtual coordinate) after displacement of the virtual particle. For example, if "B" in the character string "D inputs A to B and determines C" in the natural text to be analyzed is displaced by more than a predetermined criterion in the dynamics calculation, the search unit 30 acquires the vector (virtual coordinate) of "B" and searches the element group (e.g., a patent specification) for a replacement candidate having a vector (virtual coordinate) that is closest to the vector (virtual coordinate), thereby obtaining a replacement candidate "X". After that, the process proceeds to step S411.

[0101] In step S411, the displaced element is replaced with the searched replacement candidate. The correction unit 31 corrects (replaces) the element corresponding to the displaced virtual particle in the natural sentence to be analyzed, using the element searched by the search unit 30. For example, when "X" is searched out from the element group as a replacement candidate for "B" in the character string "D inputs A into B and determines C" in the natural sentence to be analyzed, the correction unit 31 replaces "B" with "X" to correct the natural sentence to be analyzed to "D inputs A into X and determines C". This correction may be performed automatically or after confirmation and approval by the user. After that, the process shown in this flowchart ends.

[0102] In the above description, the natural text to be analyzed is a patent claim, but the application of the present disclosure is not limited to patent claims. By setting the element group for searching for replacement candidates as a feature space according to the attributes of the natural text to be analyzed, various types of natural text to be analyzed can be corrected. For example, when the natural text to be analyzed is an advertising copy, the element group for searching for replacement candidates may be some document, dictionary, database, Internet, etc. narrowed down to those related to advertising. In this way, for example, when it is determined that the displacement of the element "full" in the advertising copy "you can get a lot of points" is equal to or greater than a predetermined standard (unstable), it is possible to search for a replacement candidate "plenty" from a database of terms related to advertising, and correct the advertising copy to "you can get a lot of points."

[0103] According to the system of the present embodiment described above, it is possible to quantitatively evaluate natural text, and further to search for replacement candidates for elements related to displaced virtual particles and replace elements as necessary, thereby improving the evaluation value of the natural text to be analyzed. [Explanation of symbols]

[0104] 1, 1b, 1c, 1d Information processing device

Claims

1. a coordinate setting means for setting virtual coordinates of each of a plurality of elements constituting the natural text to be analyzed based on a distributed representation of the element; a mass setting means for regarding the plurality of elements as a plurality of virtual particles placed at the virtual coordinates corresponding to each element, and setting virtual masses for the plurality of virtual particles; a displacement calculation means for calculating displacements of the plurality of virtual particles by applying a potential function to the plurality of virtual particles and performing a dynamics calculation; an evaluation value calculation means for calculating an evaluation value indicating an interaction between the plurality of elements constituting the natural text to be analyzed based on the displacements calculated for the plurality of virtual particles; A natural language processing system that utilizes dynamics calculations.

2. the mass setting means sets virtual masses for the plurality of virtual particles according to the abstraction levels of the elements corresponding to the virtual particles. A natural language processing system utilizing the dynamics calculation according to claim 1.

3. a fixing means for fixing at least one virtual particle among the plurality of virtual particles so that the fixed virtual particle is not displaced in the dynamics calculation by the displacement calculation means; a searching means for searching for an element that is similar to or coincides with the vector or virtual coordinates after the displacement of the virtual particle by searching a group of elements different from the natural text to be analyzed; The natural language processing system using dynamics calculation according to claim 1 , further comprising:

4. The method further includes a correction unit that corrects an element corresponding to the virtual particle after the displacement in the natural sentence to be analyzed by using the element retrieved by the search unit. A natural language processing system utilizing the dynamics calculation according to claim 3.

5. the fixing means identifies an element to be fixed by analyzing a relationship between elements, and fixes at least a virtual particle corresponding to the element to be fixed among the plurality of virtual particles, thereby preventing the fixed virtual particle from being displaced in the dynamics calculation by the displacement calculation means. A natural language processing system utilizing the dynamics calculation according to claim 3.

6. the fixing means identifies elements that are not case components by analyzing a case dominance relationship of each element, and fixes at least virtual particles corresponding to the elements that are not case components among the plurality of virtual particles, thereby preventing the fixed virtual particles from being displaced in the dynamics calculation by the displacement calculation means; A natural language processing system utilizing the dynamics calculation according to claim 5.

7. an element specifying means for extracting a plurality of morphemes constituting the analysis target natural sentence by performing a morpheme analysis of the analysis target natural sentence, and specifying the plurality of morphemes as the plurality of elements; A natural language processing system utilizing the dynamics calculation according to claim 1.

8. A unit dividing means for dividing a natural text to be analyzed into a plurality of text units; an element coordinate setting means for setting virtual coordinates of a virtual particle when the element is regarded as a virtual particle for each of a plurality of elements constituting each of the plurality of sentence units, based on a distributed representation of the element corresponding to the virtual particle; an inter-element energy calculation means for calculating, for each of the plurality of sentence units, energy indicating an interaction between the virtual particles corresponding to the plurality of elements by regarding the plurality of elements constituting the target sentence unit as a plurality of virtual particles placed at the virtual coordinates corresponding to each element and performing a force field calculation by applying a potential function to the plurality of virtual particles; a unit coordinate setting means for setting virtual coordinates of each of the plurality of sentence units constituting the natural sentence to be analyzed, the virtual coordinates of the sentence unit being based on the virtual coordinates corresponding to the plurality of elements constituting the sentence unit; a mass setting means for regarding the plurality of text units as a plurality of virtual particles placed at the virtual coordinates corresponding to each text unit, and setting a virtual mass for each of the plurality of text units based on the energy indicating an interaction between the plurality of virtual particles corresponding to the plurality of elements constituting a target text unit; a displacement calculation means for calculating displacements of the virtual particles corresponding to the plurality of sentence units by applying a potential function to the virtual particles corresponding to the plurality of sentence units and performing a dynamics calculation; an evaluation value calculation means for calculating an evaluation value indicating an interaction between the plurality of sentence units constituting the analysis target natural sentence based on the displacements calculated for the plurality of virtual particles; A natural language processing system that utilizes dynamics calculations.

9. a parameter setting unit that sets parameters for the dynamics calculation in accordance with environmental data related to the natural text to be analyzed; A natural language processing system utilizing the dynamics calculation according to claim 1 or 8.

10. On the computer, a coordinate setting step of setting virtual coordinates of each of a plurality of elements constituting the natural text to be analyzed based on a distributed representation of the element; a mass setting step of regarding the plurality of elements as a plurality of virtual particles placed at the virtual coordinates corresponding to each element and setting virtual masses for the plurality of virtual particles; a displacement calculation step of calculating displacements of the plurality of virtual particles by performing dynamics calculation by applying a potential function to the plurality of virtual particles; an evaluation value calculation step of calculating an evaluation value indicating an interaction between the plurality of elements constituting the natural text to be analyzed based on the displacements calculated for the plurality of virtual particles; A natural language processing method for executing the above.

11. On the computer, A unit division step of dividing a natural text to be analyzed into a plurality of text units; an element coordinate setting step of setting virtual coordinates of a virtual particle when each of a plurality of elements constituting each of the plurality of sentence units is regarded as a virtual particle based on a distributed representation of the element corresponding to the virtual particle; an inter-element energy calculation step of regarding each of the plurality of sentence units as a plurality of virtual particles placed at the virtual coordinates corresponding to each element, and calculating an energy indicating an interaction between the virtual particles corresponding to the plurality of elements by applying a potential function to the plurality of virtual particles and performing a force field calculation; a unit coordinate setting step of setting virtual coordinates of each of the plurality of sentence units constituting the natural sentence to be analyzed, the virtual coordinates of the sentence unit being based on the virtual coordinates corresponding to the plurality of elements constituting the sentence unit; a mass setting step of regarding the plurality of text units as a plurality of virtual particles placed at the virtual coordinates corresponding to each text unit, and setting a virtual mass for each of the plurality of text units based on the energy indicating an interaction between the plurality of virtual particles corresponding to the plurality of elements constituting a target text unit; a displacement calculation step of calculating displacements of the virtual particles corresponding to the sentence units by applying a potential function to the virtual particles corresponding to the sentence units and performing dynamics calculation; an evaluation value calculation step of calculating an evaluation value indicating an interaction between the plurality of sentence units constituting the analysis target natural sentence based on the displacements calculated for the plurality of virtual particles; A natural language processing method for executing the above.

12. Computer, a coordinate setting means for setting virtual coordinates of each of a plurality of elements constituting the natural text to be analyzed based on a distributed representation of the element; a mass setting means for regarding the plurality of elements as a plurality of virtual particles placed at the virtual coordinates corresponding to each element, and setting virtual masses for the plurality of virtual particles; a displacement calculation means for calculating displacements of the plurality of virtual particles by applying a potential function to the plurality of virtual particles and performing a dynamics calculation; an evaluation value calculation means for calculating an evaluation value indicating an interaction between the plurality of elements constituting the natural text to be analyzed based on the displacements calculated for the plurality of virtual particles; A program that functions as a

13. Computer, A unit dividing means for dividing a natural text to be analyzed into a plurality of text units; an element coordinate setting means for setting virtual coordinates of a virtual particle when the element is regarded as a virtual particle for each of a plurality of elements constituting each of the plurality of sentence units, based on a distributed representation of the element corresponding to the virtual particle; an inter-element energy calculation means for calculating, for each of the plurality of sentence units, energy indicating an interaction between the virtual particles corresponding to the plurality of elements by regarding the plurality of elements constituting the target sentence unit as a plurality of virtual particles placed at the virtual coordinates corresponding to each element and performing a force field calculation by applying a potential function to the plurality of virtual particles; a unit coordinate setting means for setting virtual coordinates of each of the plurality of sentence units constituting the natural sentence to be analyzed, the virtual coordinates of the sentence unit being based on the virtual coordinates corresponding to the plurality of elements constituting the sentence unit; a mass setting means for regarding the plurality of text units as a plurality of virtual particles placed at the virtual coordinates corresponding to each text unit, and setting a virtual mass for each of the plurality of text units based on the energy indicating an interaction between the plurality of virtual particles corresponding to the plurality of elements constituting a target text unit; a displacement calculation means for calculating displacements of the virtual particles corresponding to the sentence units by applying a potential function to the virtual particles corresponding to the sentence units and performing a dynamics calculation; an evaluation value calculation means for calculating an evaluation value indicating an interaction between the plurality of sentence units constituting the analysis target natural text based on the displacements calculated for the plurality of virtual particles; A program that functions as a

Images