KR102085217B1 - Method, apparatus and system for determining similarity of patent documents - Google Patents

Abstract

Provided are a method, an apparatus, and a system for determining a similarity of patent documents. A control method of the apparatus for determining a similarity of patent documents comprises the following steps: obtaining, by a server, a target patent document; obtaining, by the server, at least one word on the basis of the target patent document; obtaining, by the server, an importance score of the obtained at least one word; obtaining, by the server, a similar patent document by inputting the target patent document to a first artificial intelligence model; obtaining, by the server, a plurality of sentences included in the target patent document and a plurality of sentences included in the similar patent document; obtaining, by the server, a first sentence among the plurality of sentences included in the target patent document and a second sentence among the plurality of sentences included in the similar patent document; obtaining, by the server, evaluation results with respect to the first sentence and the second sentence by inputting the first sentence and the second sentence to a second artificial intelligence module; and generating, by the server, a prior art research report of the target patent document on the basis of the evaluation results.

Description

Method, apparatus and system for determining similarity of patent document {METHOD, APPARATUS AND SYSTEM FOR DETERMINING SIMILARITY OF PATENT DOCUMENTS}

The present invention relates to a method, apparatus and system for determining similarity of a patent document.

With the development of the fourth industry, the value of intellectual property rights is increasing. Accordingly, many people are trying to protect their technology, acquire rights to the technology, and interest in patent applications for the technology is increasing.

On the other hand, for patent applications, prior art research is performed to determine whether a technology can be patented, and prior art research may be performed by searching various patent documents published in the past.

The most important thing in carrying out the prior art research is to determine whether there is a similar patent document that can deny the progress of the target patent document.

However, due to the huge amount of patent documents and due to time constraints, it is virtually impossible to analyze all published patent documents, and in order to obtain the maximum results within a given time, a search method or the like is provided. It is a reality to obtain a similar patent document.

However, the acquisition of similar patent documents through search input or the like is often dependent on the ability of a manpower to perform prior art research, and thus it is often not guaranteed for stable results for prior art research.

Therefore, there is a need for a prior art research method that can guarantee stable results with little time investment.

Patent Publication No. 10-2018-0110713, Oct. 11, 2018

The problem to be solved by the present invention is to provide a method, apparatus and system for determining similarity of patent documents.

Problems to be solved by the present invention are not limited to the above-mentioned problems, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method of controlling a similarity determination system of a patent document, the method including: obtaining, by a server, a target patent document; Acquiring, by the server, at least one word based on the target patent document; Acquiring, by the server, an importance score of the obtained at least one word; Obtaining, by the server, a similar patent document by inputting the target patent document into a first artificial intelligence model; Acquiring, by the server, a plurality of sentences included in the target patent document, and obtaining a plurality of sentences included in the similar patent document; Acquiring, by the server, a first sentence among a plurality of sentences included in the target patent document and a second sentence among a plurality of sentences included in the similar patent document; Acquiring, by the server, the first sentence and the second sentence into a second artificial intelligence model to obtain evaluation results of the first sentence and the second sentence; And generating, by the server, a prior art research report on the target patent document based on the evaluation result.

In this case, the acquiring the word may include: acquiring, by the server, a plurality of words by analyzing a morpheme based on the target patent document; Determining, by the server, an error word among the obtained plurality of words; The server correcting the error word; And obtaining, by the server, at least one compound noun phrase set based on the degree of association between the plurality of words.

In this case, the acquiring the importance score may include: acquiring, by the server, a word that is a calculation target of the importance score from the target patent document; The server includes the first detail importance of the word in the entire patent document, the second detail importance of the word in the patent classification information corresponding to the technical field information of the target patent document, and the word among the entire patent document. Calculating at least one detail importance of the third detail importance of the searched patent document; And calculating, by the server, the importance score of the word based on one or more of the first detail importance, the second detail importance and the third detail importance.

At this time, the control method, the server, the step of obtaining the entire patent document that is the basis of the thesaurus; Analyzing, by the server, morphemes for each of the entire patent documents; Converting a word included in each of the entire patent documents into a word vector based on the result of the morphological analysis; Calculating, by the server, the similarity between the word vectors; And grouping, by the server, a word corresponding to the word vector into a group of similar words based on the similarity.

In this case, the determining of the error word may include: matching semantic information on each of the obtained plurality of words; Obtaining an accuracy score for each of the plurality of words based on the matched semantic information and the target patent document; And determining a word having an accuracy score equal to or less than a preset value as an error word, and correcting the error word comprises: obtaining a plurality of semantic information about the error word; Obtaining a plurality of weights for each of the obtained plurality of semantic informations; Obtaining a weight having the highest association with the target patent document among the plurality of weights; And matching the semantic information corresponding to the weight having the highest association with the target patent document to the error word, and obtaining the at least one compound noun phrase set includes: Obtaining a plurality of words for a sentence; Obtaining a plurality of compound noun phrase set candidates obtained by combining the plurality of words; Obtaining a frequency at which a compound noun phrase set identical to the obtained plurality of compound noun phrase set candidates is included in the target patent document; Determining a compound noun phrase set candidate whose frequency is equal to or greater than a preset frequency as a compound noun phrase set; Wherein the compound noun phrase set candidate includes order information and spacing information of words included in the compound noun phrase set candidate.

At this time, the step of calculating the detailed importance of the word, the first appearance ratio of the number of appearance of the word in the entire patent document relative to the total number of words of the entire patent document and the total of the total number of sentences of the entire patent document Calculating the first sub-critical importance based on a second appearance ratio of the number of sentences in which the word appears in sentences of a patent document; The third occurrence rate of the number of occurrences of the word in the patent classification information to the total number of words of the patent classification information and the number of appearance sentences in which the word appeared among the sentences of the entire patent document compared to the total number of sentences of the entire patent document Calculating the second sub-critical importance based on a fourth appearance ratio of the second; And calculating an influence value of each of the search patent documents based on reference information of each of the search patent documents, and calculating a third detailed importance of the search patent document using the influence value. The calculating of the detailed importance may include calculating the first detail importance using Equation 1 below, and calculating the second detail importance may calculate the second detail importance using Equation 1 below. It includes a step.

<Equation 1>

<Equation 2>

Here, the W1 is the first detail importance, the wpw is the number of occurrences of the word in the entire patent document, the WPW is the total number of words in the patent document, the wps is the The word is the number of appearance sentences appeared, the WPS is the total number of sentences of the patent document, the a1 is the control constant of the second appearance ratio, the W2 is the second detail importance, the ipcw is the patent classification The number of occurrences of the word in the information, the IPCW is the total number of words in the patent classification information, the ipcs is the number of appearance sentences where the word appeared in the sentences of the entire patent document, the IPCS is A2 is an adjustment constant of the fourth occurrence rate.

In this case, the obtaining of the similar patent document may include: obtaining a plurality of words included in the target patent document; Clustering the obtained plurality of words to obtain a plurality of target patent clusters; Acquiring a plurality of midpoints for each of the plurality of target patent clusters, and determining a position of the target patent document based on the plurality of midpoints for each of the plurality of target patent clusters and the number of words included in the target patent cluster. Making; Obtaining a plurality of words included in a patent document and clustering the obtained plurality of words to obtain a plurality of patent clusters; Acquiring a plurality of midpoints for each of the plurality of patent clusters, and determining a position of the patent document based on the plurality of midpoints for each of the plurality of patent clusters and the number of words included in the patent cluster; And determining the patent document as the similar patent document when the position of the target patent document and the position of the patent document are within a preset distance, and an evaluation result of the first sentence and the second sentence. The obtaining may include obtaining similarity scores and dissimilarity scores of the first sentence and the second sentence, respectively; Determining that the first sentence and the second sentence are irrelevant sentences when the dissimilarity score is equal to or greater than a predetermined score; Obtaining at least one word not included in the second sentence among words included in the first sentence when the similarity score is equal to or greater than a predetermined score; Obtaining at least importance scores for each of at least one word not included in the second sentence, and determining whether there is a word equal to or greater than a predetermined importance score among the obtained at least one importance scores; Determining that the first sentence and the second sentence are matched sentences when a word having a predetermined importance score or more does not exist; And determining that the first sentence and the second sentence are inconsistent sentences when a word having a predetermined importance score or more exists.

In this case, the converting into the word vector includes converting the word into a word vector through Word2Vec learning based on the patent document, and calculating the similarity between the two word vectors among the word vectors. Calculating a distance and calculating the calculated distance as a similarity, and grouping into the similar word group includes: checking whether the similarity between any two word vectors among the word vectors is less than a predetermined reference similarity, If the similarity between any two word vectors of the word vectors is less than a preset reference similarity, grouping two words corresponding to the two word vectors into the similar word group, and acquiring the entire patent document comprises: The patent document of any of the acquired patent documents Determine whether the group, and removing the patent document that meets the conditions, the noise document.

Other specific details of the invention are included in the detailed description and drawings.

According to the above-described embodiment of the present invention, the user can be provided with a prior art research method that can ensure a stable result with little time investment.

Effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a system diagram according to an embodiment of the present invention.
2 is a flowchart illustrating a method of obtaining a prior art research report according to an embodiment of the present invention.
3 is a flowchart illustrating a method of obtaining a word according to an embodiment of the present invention.
4 is a flowchart illustrating a method of calculating importance scores according to an embodiment of the present invention.
5 is a flowchart illustrating a method of generating a thesaurus according to an embodiment of the present invention.
6A through 6C are flowcharts illustrating a word acquisition method in detail according to an embodiment of the present invention.
7 is a flowchart illustrating a method of calculating importance scores according to an embodiment of the present invention in detail.
8A and 8B are flowcharts illustrating a method for obtaining a similar patent document and a method for determining a similar sentence according to an embodiment of the present invention.
9 is a flowchart illustrating a specific method of generating a thesaurus according to an embodiment of the present invention.
10 is an exemplary diagram for describing a process of deriving an evaluation result using the first AI model and the second AI model according to an embodiment of the present invention.
11 is a block diagram of an apparatus according to an embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will be apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be embodied in various different forms, only the present embodiments are intended to complete the disclosure of the present invention, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the invention, which is defined only by the scope of the claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and / or “comprising” does not exclude the presence or addition of one or more other components in addition to the mentioned components. Like reference numerals refer to like elements throughout, and "and / or" includes each and all combinations of one or more of the mentioned components. Although "first", "second", etc. are used to describe various components, these components are of course not limited by these terms. These terms are only used to distinguish one component from another. Therefore, of course, the first component mentioned below may be the second component within the technical spirit of the present invention.

Unless otherwise defined, all terms used in the present specification (including technical and scientific terms) may be used in a sense that can be commonly understood by those skilled in the art. In addition, terms that are defined in a commonly used dictionary are not ideally or excessively interpreted unless they are specifically defined clearly.

The term "part" or "module" as used herein refers to a hardware component such as software, FPGA or ASIC, and the "part" or "module" plays certain roles. However, "part" or "module" is not meant to be limited to software or hardware. The “unit” or “module” may be configured to be in an addressable storage medium or may be configured to play one or more processors. Thus, as an example, a "part" or "module" may include components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, Procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. Functions provided within components and "parts" or "modules" may be combined into smaller numbers of components and "parts" or "modules" or into additional components and "parts" or "modules". Can be further separated.

The spatially relative terms " below ", " beneath ", " lower ", " above ", " upper " It can be used to easily describe a component and its correlation with other components. Spatially relative terms are to be understood as including terms in different directions of components in use or operation in addition to the directions shown in the figures. For example, when flipping a component shown in the drawing, a component described as "below" or "beneath" of another component may be placed "above" the other component. Can be. Thus, the exemplary term "below" can encompass both an orientation of above and below. Components may be oriented in other directions as well, so spatially relative terms may be interpreted according to orientation.

In the present specification, a computer refers to any kind of hardware device including at least one processor, and according to an embodiment, it may be understood as a meaning encompassing a software configuration that operates on the hardware device. For example, a computer may be understood as including, but not limited to, a smartphone, a tablet PC, a desktop, a notebook, and a user client and an application running on each device.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention;

Each step described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed in different devices according to embodiments.

1 is a system diagram according to an embodiment of the present invention.

The patent document similarity determination system according to the present invention includes a server 10 and an electronic device 20.

The server 10 is configured to obtain a target patent document, obtain a similar patent document from the obtained target patent document, and determine the similarity between sentences of the target patent document and sentences of the similar patent document.

In an embodiment, the server 10 may receive a target patent document from the electronic device 20 or obtain a target patent document from an external server.

In another embodiment, the server 10 may obtain a plurality of words from the target patent document or the similar patent document, and obtain the importance score of the acquired words. In addition, the server 10 may obtain a thesaurus from the entire patent document and use it to determine the importance score of the word.

In another embodiment, the server 10 may obtain a similar patent document for the target patent document, and obtain a similar sentence to generate a prior art research report.

In the present specification, a patent document is a concept including a target patent document and a similar patent document, and may be a document on a technical content submitted by an applicant for patent registration at each patent office of each country. However, the present invention is not limited thereto, and the patent document may be understood as a concept including various documents including technical contents such as job inventions and papers for patent application. According to an embodiment, the target patent document may be at least one of a job invention and a paper for a patent application, and the similar patent document may be at least one of a job invention, a paper and a patent application for a patent application.

The electronic device 20 is a component for providing a patent document to the server 10. The electronic device 200 according to the present invention may be implemented as a smart phone, but this is only an example, and may include a smart phone, a tablet personal computer, a mobile phone, a video phone, E-book readers, desktop PCs, laptop PCs, netbook computers, workstations, servers, personal digital assistants, portable multimedia It may include at least one of a player or a wearable device.

2 is a flowchart illustrating a method of obtaining a prior art research report according to an embodiment of the present invention.

In operation S110, the server 10 may acquire a target patent document.

In operation S120, the server 10 may obtain at least one word based on the target patent document.

In operation S130, the server 10 may acquire an importance score of the obtained at least one word.

In operation S140, the server 10 may obtain a similar patent document by inputting the target patent document into the first artificial intelligence model.

In detail, the server 10 may obtain a plurality of words from the target patent document or the similar patent document, and obtain the importance score of the acquired words. In addition, the server 10 may obtain a thesaurus from the entire patent document and use it to determine the importance score of the word.

In one embodiment, the server 10, based on the morphological analysis of each of the plurality of patent documents stored in the first database, analyzes each of the plurality of patent documents, the target patent based on the morphological analysis of the target patent document After analyzing the document, a similar patent document highly related to the target patent document can be obtained.

In operation S150, the server 10 may obtain a plurality of sentences included in the target patent document and may obtain a plurality of sentences included in the similar patent document.

In operation S160, the server 10 may acquire a first sentence of the plurality of sentences included in the target patent document, and obtain a second sentence of the plurality of sentences included in the similar patent document.

In operation S170, the server 10 may obtain a result of evaluating the first sentence and the second sentence by inputting the first sentence and the second sentence into the second artificial intelligence model.

In operation S180, the server 10 may generate a prior art research report on the target patent document based on the evaluation result. Specifically, the prior art research report on the target patent document may be generated based on the evaluation results of the first sentence and the second sentence. In one embodiment, when the evaluation results for the first sentence and the second sentence is obtained in the form of a score, the server 10 is the first sentence of which the evaluation result score for the first sentence and the second sentence is greater than or equal to the predetermined score. And a prior art research report based on the second sentence corresponding to the first sentence.

Meanwhile, the first artificial intelligence model is an artificial intelligence model based on a convolutional deep neural network (CNN) trained by inputting a plurality of patent documents as learning data, and the second artificial intelligence model includes a plurality of patent documents and the It may be an AI model based on a Bi-LSTM model trained on the basis of two patent documents among a plurality of patent documents and a prior art research report on the two patent documents. However, the present invention is not limited thereto, and various artificial intelligence models may be applied to the present invention. For example, a model such as a deep neural network (DNN), a recurrent neural network (RNN), and a bidirectional recurrent deep neural network (BRDNN) may be used as an artificial intelligence model, but is not limited thereto.

In this case, convolutional deep neural networks (CNNs) are a type of multilayer perceptrons designed to use minimal preprocessing. A convolutional neural network consists of one or several convolutional layers and a general artificial neural network layer on top of them, and further utilizes weights and pooling layers. This structure allows the multiplicative neural network to fully utilize the input data of the two-dimensional structure. In addition, convolutional neural networks can be trained through standard backtransmission. The multiplicative neural network is more easily trained than other feedforward neural network techniques and has the advantage of using fewer parameters. The convolutional neural network extracts features from the input image by performing alternating convolutional and subsampling on the input image.

The convolutional neural network includes a plurality of convolution layers, a plurality of subsampling layers, a lacal pooling layer, a max-pooling layer, and a fully connected layer. The convolution product layer is a layer that performs a convolution product on an input image. The subsampling layer is a layer that locally extracts the maximum value of the input image and maps it to the 2D image. The subsampling layer enlarges the local region and performs subsampling.

In the product multiplication layer, information such as kernel size, the number of kernels to be used (that is, the number of maps to be generated), and a weight table to be applied when performing a product multiplication are needed. For example, assume that the size of the input image is 32 × 32, the size of the kernel is 5 × 5, and the number of kernels to be used is 20. In this case, if a 5 × 5 kernel is applied to an input image having a size of 32 × 32, it is impossible to apply the kernel to two pixels at the top, bottom, left, and right sides of the input image. When the kernel is placed on the input image and then the result is composite, the resulting value of '-8' is determined as the pixel value corresponding to the center element of the kernel among the pixels of the input image included in the kernel. to be. Therefore, when a composite product is performed by applying a 5 × 5 kernel to a 32 × 32 input image, a 28 × 28 map is generated. Since it is assumed that the total number of kernels to be used is 20 in total, 20 maps of 28 × 28 sizes are generated in the first convolutional layer.

In the subsampling layer, information on the size of the kernel to be subsampled and information on whether to select the maximum value or the minimum value among the values in the kernel region are needed.

Also, deep neural networks (DNNs) are artificial neural networks (ANNs) including a plurality of hidden layers between an input layer and an output layer.

At this time, the structure of the deep neural network may be composed of a perceptron (perceptron). A perceptron consists of several inputs, one processor, and one output. The processor multiplies a plurality of input values by each weight, and then adds the weighted input values together. The processor then substitutes the sum of the values into the activation function and outputs one output value. If a specific value is desired as an output value of the activation function, the weight multiplied by each input value may be modified, and the output value may be recalculated using the modified weight. At this time, each perceptron may use a different activation function. Each perceptron also accepts the output from the previous layer as input, then uses the activation function to get the output. The obtained output is passed to the input of the next layer. Through the above process, several output values can be finally obtained.

In addition, returning to the deep learning technique, a cyclic neural network (RNN) refers to a neural network in which a connection between units forming an artificial neural network forms a directed cycle. The cyclic neural network, unlike the front neural network, can utilize the memory inside the neural network to process arbitrary input.

Deep Belief Networks (DBN) is a generic graphical model used in machine learning. In deep learning, deep Belief Networks (DBN) is a deep neural network composed of multiple layers of latent variables. There is a connection between layers, but there is no connection between units within a layer.

Deep trust neural networks can be used for preliminary learning due to the nature of the generation model, and after the initial weights are learned through prior learning, the weights can be fine-tuned through back propagation or other discrimination algorithms. This characteristic is very useful when the training data is small, because the smaller the training data, the greater the influence of the initial value of the weight on the resulting model. The pre-learned weight initial value is closer to the optimal weight than the arbitrarily set weight initial value, which enables the performance and speed improvement of the fine tuning step.

Details of the above-described artificial intelligence and its learning method are described for the purpose of illustration, and the artificial intelligence used in the embodiments described below and its learning method are not limited. For example, all kinds of artificial intelligence techniques and learning methods that can be applied by those skilled in the art to solve the same problem may be utilized to implement a system according to the disclosed embodiments.

3 is a flowchart illustrating a method of obtaining a word according to an embodiment of the present invention.

In operation S210, the server 10 may obtain a plurality of words by analyzing a morpheme based on the target patent document.

In one embodiment, the server 10 may perform a morphological analysis of the target patent document using the Mecab morpheme analyzer. However, the present invention is not limited thereto, and various morpheme analyzers such as Okt, Komoran, Hannanum, and Kkma morpheme analyzers may be used. Furthermore, the server 10 may use various morpheme analyzers according to the language of the target patent document to be analyzed. On the other hand, the present invention has been described a method for analyzing the morphemes of the target patent document, the server 10, of course, can perform a morphological analysis of various patent documents as well as the target patent document.

In operation S220, the server 10 may determine an error word among the obtained plurality of words.

In an embodiment, if the acquired plurality of words does not satisfy a preset condition, the server 10 may determine the corresponding word as an error word.

In operation S230, the server 10 may correct an error word.

In operation S240, the server 10 may obtain at least one compound noun phrase set based on the degree of association between the plurality of words.

4 is a flowchart illustrating a method of calculating importance scores according to an embodiment of the present invention.

In operation S310, the server 10 may obtain a word, which is a calculation target of the importance score, from the target patent document.

In step S320, the server 10 determines that the first detailed importance of the words in the entire patent document, the second detailed importance of the words in the patent classification information corresponding to the technical field information of the target patent document, and the words in the entire patent document are displayed. At least one detail importance of the third detail importance of the included search patent document may be calculated.

In operation S330, the server 10 may calculate an importance score of the word based on at least one of the first detail importance level, the second detail importance level, and the third detail importance level.

Specifically, the server 10 may extract only nouns by performing morphological analysis on the target patent document. In this case, it is noted that the type of the morphological analysis method is not limited as long as the server 10 extracts a noun from the target patent document.

Subsequently, the server 10 may determine the technical field information of the target patent document based on the extracted noun. At this time, the server 10 may compare the pre-stored word data for each technical field information and the extracted noun, and as a result of the comparison, determine the technical field information including the extracted noun as the technical field information of the target patent document.

Thereafter, the server 10 may exclude a word set as a stop word among the words included in the target patent document from the word that is the target of calculating the importance score. Specifically, the server 10 reads stopword data for each technical field information corresponding to the technical field information of the target patent document, and includes words included in the extracted terminology data for each technical field information among the words extracted from the target patent document. May be excluded from the words to be calculated for the importance score.

Here, the stop word may mean a word that is frequently used in the related art but does not have a technical meaning. For example, computers are frequently used in the field of IT technology but do not have a technical meaning related to IT technology and may be defined as a stopword.

Accordingly, the server 10 may obtain a word that is not a stopword among the words extracted from the target patent document as a word that is a calculation target of the importance score.

5 is a flowchart illustrating a method of generating a thesaurus according to an embodiment of the present invention.

In operation S410, the server 10 may acquire the entire patent document that is the basis of the thesaurus. In detail, the server 10 may obtain a patent document that is disclosed as time passes in real time to obtain the entire patent document. In this case, the server 10 may receive the latest published patent document from the external server or the electronic device 20 through communication and store it in the memory 104.

In one embodiment, the server 10 may check whether any one of the obtained patent documents meet the noise document conditions. Specifically, the server 10 may remove the patent document that satisfies the noise document condition from the entire patent document. Here, the noise document condition is to check whether the patent document is a noise document. If the patent document has a space equal to or greater than a preset reference size in each patent document, and if the same character is repeated more than a preset number of times, the text in the entire patent document It is the case that the ratio with the image is more than the preset reference ratio.

That is, the server 10 may check whether or not there is noise for all patent documents. The server 10 may remove the noise document that satisfies the noise document condition.

In operation S420, the server 10 may analyze the morpheme for each patent document.

In operation S430, the server 10 may convert a word included in each patent document into a word vector based on the result of the morpheme analysis.

In this case, the server 10 may project the plurality of word characteristics of the above-described words into a multidimensional real space and vectorize them to be converted into a word vector. In one embodiment, server 10 may convert a word vector to a word vector using Word2vec learning.

In addition, the server 10 may express a word in a vector space of about 200 to 300 dimensions, and CBOW (Continuous Bag of Words) and one word predicting a target word based on the direction of meaning created by surrounding words for learning. You can use the Skip-gram model to predict words that can come around.

In this case, the distance between two word vectors may indicate similarity between words corresponding to each of the two word vectors, and the direction of the word vector may indicate meaning in the patent document.

In operation S440, the server 10 may calculate the similarity between the word vectors.

In detail, the server 10 may calculate the cosine similarity between two word vectors or normalize the cosine similarity to calculate the similarity between two word vectors.

For example, the server 10 may calculate the degree of similarity between two word vectors using a cosine of an angle between two word vectors in real space. In addition, the server 10 may calculate the degree of similarity between two word vectors by normalizing a cosine similarity value between two word vectors to have a range of 0 to 1.

That is, the server 10 may convert all words included in the entire patent radio into a word vector, calculate a distance between all the converted word vectors with cosine similarity, and calculate the calculated distance with similarity.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may perform a post-processing process of the thesaurus dictionary generation method of the patent document.

In detail, the server 10 may determine whether a word corresponding to the word vector is included in the similar word group based on the similarity. Specifically, the server 10 determines whether the similarity between any two word vectors is less than the preset reference similarity, and if the similarity between any two word vectors is less than the preset reference similarity, the words corresponding to the two word vectors are the same similar words. It can be confirmed that it is included in the group.

Thereafter, the server 10 may group the words corresponding to the word vectors into similar word groups based on the similarity.

In detail, when it is confirmed that the words corresponding to the two word vectors are included in the same similar word group, the server 10 may include the two words in the same similar word group.

Meanwhile, even when the similarity between any two word vectors is less than the preset reference similarity, the server 10 may determine that the similarity between the word vectors corresponding to the words previously included in the similar group exceeds the preset maximum similarity. Words corresponding to two word vectors less than the similarity may not be included in the similar word group.

Through this, the server 10 may prevent the similarity group from being expanded to reduce the similarity between any two words.

6A through 6C are flowcharts illustrating a word acquisition method in detail according to an embodiment of the present invention.

In detail, as illustrated in FIG. 6A, in operation S505, the server 10 may match semantic information about each of the acquired words.

In this case, the semantic information may mean an intent for a word. The server 10 may perform natural language processing using an artificial intelligence model to obtain semantic information about a word.

Specifically, the artificial intelligence model includes a natural language understanding unit, the natural language understanding unit may grasp the intent of the words included in the entity and the sentence based on the sentence analysis result, and furthermore, the natural language understanding unit may structure the sentence. And sentence analysis through the analysis of the principal components and statistics / analysis can be performed using the sentence.

In one embodiment, the server 10 may obtain the semantic information about the apple by analyzing a sentence including the 'apple'. For example, it may be assumed that a word acquired through sentence sentence is “apple.” In this case, semantic information about an apple may be semantic information indicating a kind of fruit as a noun, but as a verb to another person. The server 10 may match semantic information, which is determined to be suitable with a sentence, among the plurality of semantic information about the apple.

In operation S510, the server 10 may obtain an accuracy score for each of the plurality of words based on the matched semantic information and the target patent document.

In this case, the server 10 obtains semantic information about the word acquired through one sentence, but the semantic information obtained through this may be inaccurate. Therefore, the server 10 may obtain the accuracy score of the semantic information matched to the word based on the entire patent document including the sentence.

Therefore, the server 10 may obtain the accuracy score for the semantic information of the word obtained through sentence analysis based on the semantic information on the same word found in the entire patent document.

For example, the semantic information obtained by analyzing a sentence containing 'apple' is information about an apple representing fruit, but the semantic information of 'apple' found in the patent document including the sentence is different. In the case of information on telling a wrong person, the server 10 may set a low matched semantic information.

In operation S515, the server 10 may determine a word having an accuracy score equal to or less than a preset value as an error word.

That is, when the accuracy score is less than or equal to a preset value, the server 10 may determine that the semantic information matched with the corresponding word is incorrectly matched, and may acquire the word as an error word.

On the other hand, as shown in Figure 6b, in step S520, the server 10 may obtain a plurality of semantic information for the error word. In detail, the server 10 may obtain a plurality of semantic information of an error word from a plurality of patent documents.

In operation S525, the server 10 may obtain a plurality of weights for each of the obtained plurality of semantic informations.

In one embodiment, the server 10 may obtain a weight based on the target patent document including the error word. In detail, the server 10 may obtain at least one sentence including an error word in the entire patent document, and obtain semantic information on the same word as the error word included in the obtained at least one sentence. The server 10 may obtain weights for a plurality of semantic information based on semantic information about at least one word that is the same as an error word.

In operation S530, the server 10 may obtain a weight having the highest association with the target patent document among the plurality of weights.

In operation S535, the server 10 may match semantic information corresponding to a weight having the highest association with the target patent document to an error word. That is, the server 10 may determine the semantic information having the largest weight among the plurality of semantic information of the error word as semantic information of the error word, and correct the error word.

Meanwhile, according to various embodiments of the present disclosure, the error word may be a case in which the word is incorrectly parsed as an error in the morpheme analysis process as well as the case where the wrong semantic information is matched with the word. In one embodiment, when the sentence for acquiring the word is "natural language processing using machine learning", the server 10 may obtain a machine, running, use, natural language, processing as a word.

In this case, due to an error in morpheme analysis, a case of acquiring words such as machine, running, using, natural language, and processing may occur. In this case, the server 10 may determine 'use' as an error word and correct it.

In detail, the server 10 may determine whether the word is an error word based on the degree of association between the semantic information and the target patent document after acquiring semantic information about each acquired word. According to an embodiment, when the acquired word is found in a target patent document including a word having a predetermined frequency or more, the server 10 may determine that the word is an error-free word. In another embodiment, when a word is found more than a predetermined frequency in the entire patent document, the server 10 may determine that the word is not an error word by determining that the word is a commonly used word. In another embodiment, when the server 10 does not find semantic information about a word, the server 10 may determine the word as an error word. In another embodiment, if the semantic information matched with the word is heterogeneous with the target patent document including the word, the server 10 may determine the word as an error word.

In detail, the server 10 may acquire words included in the entire target patent document, and match and store semantic information on each of the acquired words. The server 10 may cluster a matched word and semantic information to obtain a cluster of a plurality of words having relevant semantic information. The server 10 may obtain a word cluster of a plurality of word clusters, in which a word included in the cluster is equal to or less than a preset number. In this case, the preset number may not mean the number of the plurality of identical words used in the target patent document, but may mean the number of the plurality of words having different forms. For example, even when the word "use" is found a plurality of times in the target patent document, since "use" is the same word, it can be determined as one word. The server 10 may determine the word included in the cluster obtained in the above step as an error word.

However, the present invention is not limited thereto, and the server 10 may determine, as an error word, a word included in a cluster of a plurality of word clusters having a distance from another cluster more than a preset distance. In detail, the server 10 may acquire center points of the plurality of clusters and determine the distance between clusters based on the obtained center points. For example, the obtained word cluster may be a first word cluster to a fourth word cluster. In this case, the server 10 obtains the distance between the first word cluster and the second word cluster, the distance between the first word cluster and the third word cluster, and the distance between the first word cluster and the fourth word cluster, respectively. When all three acquired distances are equal to or greater than a predetermined distance, the first word cluster may be determined as an error word.

On the other hand, as shown in Figure 6c, in step S540, the server 10 may obtain a plurality of words for the sentences included in the target patent document.

Specifically, when a compound noun phrase is formed by combining single words, the meaning may be different. Thus, there is a need for the server 10 to obtain a complex noun phrase set from a single word. Thus, the server 10 may obtain a compound noun phrase set based on a plurality of words.

For example, as a result of analyzing the sentence "to perform natural language processing using machine learning", the server 10 may acquire a word of a machine, a learning, a use, a natural language, and a processing word.

In operation S545, the server 10 may obtain a plurality of compound noun phrase set candidates obtained by combining a plurality of words.

For example, the server 10 may obtain a compound noun phrase set candidate such as machine learning, learning use, user language, natural language processing, etc. from the obtained word combination.

In operation S550, the server 10 may acquire a frequency at which the compound noun phrase set identical to the obtained plurality of compound noun phrase set candidates is included in the target patent document.

In operation S555, the server 10 may determine a compound noun phrase set candidate having a frequency equal to or greater than a preset frequency as a compound noun phrase set.

For example, the server 10 obtains information about machine learning total 301 times, learning usage once, user language total 0 times, and natural language processing 58 times, and the frequency of occurrence is set in advance. A compound noun phrase candidate set of more than a frequency can be obtained as a compound noun phrase set.

In this case, the compound noun phrase set candidate may include order information and spacing information of words included in the compound noun phrase set candidate.

In one embodiment, a compound noun phrase set is determined by a combination of two or more adjacent words, but in the present invention, a combination of words in which words are spaced may also be obtained as a compound noun phrase set. For example, when the server 10 obtains a compound noun phrase set from a sentence including "a first device of a plurality of devices, a second device of a plurality of devices, a third device of a plurality of devices", The first device of the device, the second device of the plurality of devices, and the third device of the plurality of devices can be obtained as a set of independent compound noun phrases, but a " device of a plurality of devices (spaced word 1) " Of course it can also be obtained. In this case, the complex noun phrase set may include spaced information. In the above-described embodiment, the separation information may be information indicating that one word is included between the word “in” and the word “device”. According to various embodiments, a word and a plurality of words may be included between the words.

개이고, 단어간의 순서가 변경되는 경우 획득되는 복합 명사구 세트의 수는

개이다. 따라서, 문장이 길어지면 길어질수록, 서버(10)는 복합 명사구 세트 검색을 위해 과도한 리소스를 투입하여야 한다. 따라서, 서버(10)는 복합 명사구 세트를 구성하는 단어의 최대 개수를 결정하고, 결정된 개수 내의 복합 명사구 세트를 획득할 수 있다. 이때, 서버(10)는, 단어 획득을 수행하는 대상특허문서와 동일한 기술 분야인 복수개의 특허문서를 획득하고, 획득된 특허문서에 포함된 복합 명사구 세트에 포함된 단어의 최대 개수를 복합 명사구 세트를 구성하는 단어의 최대 개수로 결정할 수 있다. In another embodiment, the server 10 may determine the maximum number of words constituting the compound noun phrase set, and obtain the compound noun phrase set within the determined number. Specifically, when n words are included in a sentence for obtaining a compound noun phrase set, and the order between words does not change, the number of compound noun phrase sets obtained is

, And the number of compound noun phrase sets obtained when the order of words changes

Dog. Therefore, the longer the sentence is, the longer the server 10 has to inject excessive resources to search for a compound noun phrase set. Accordingly, the server 10 may determine the maximum number of words constituting the compound noun phrase set, and obtain the compound noun phrase set within the determined number. In this case, the server 10 obtains a plurality of patent documents in the same technical field as the target patent document for performing word acquisition, and sets the maximum number of words included in the compound noun phrase set included in the obtained patent document. It can be determined by the maximum number of words constituting the.

On the other hand, the method of obtaining a compound noun phrase set according to the present invention is not limited to the above-described method, of course, can be obtained by various methods.

In an embodiment, when the plurality of words satisfy a preset condition, the server 10 may change the positions of the plurality of words.

In general, when trying to obtain a compound noun phrase in a sentence, it is unnecessary to change the order of the words included in the sentence. For example, in a sentence of "~" using machine learning, the server 10 needs to acquire a complex noun phrase set "machine learning", but obtains a compound noun phrase set of "treadmill" that changes the order of words. You won't have to. Rather, when a compound noun phrase set of a "treading machine" is obtained, there is a possibility that a compound noun phrase set having a completely different meaning is obtained, thereby degrading the performance to be achieved in the present invention.

However, if the predetermined condition is satisfied, the server 10 may obtain a compound noun phrase set candidate by changing the order of words included in the sentence. In this case, the preset condition may include a condition in which a plurality of words are adjacent to each other, a condition in which a predetermined code is included between a plurality of adjacent words, a condition in which a word located at the right of the plurality of adjacent words includes parentheses, and a language in which the plurality of words are different May be a condition.

For example, when the sentence to be analyzed includes "Machine Learning", the server 10 may acquire a machine, a running, a machine, and a learning word. Furthermore, the server 10 may obtain "machine learning" as a compound noun phrase set. Furthermore, the server 10 may determine that a parenthesis code, which is a preset condition, is included, and acquire “Machine Learning Machine Learning” as a compound noun phrase set. That is, the server 10 is expressed in a different language when a compound noun phrase (or word) called "machine learning" and a compound noun phrase of another language for the compound noun phrase (or word) are present together. Two compound noun phrases (or words) having a can be obtained as a compound noun phrase, and a compound noun phrase can be obtained by changing the order of the two compound noun phrases. This has the effect of increasing the efficiency of comparison when comparing patent documents written in different languages. That is, a patent document written in Korean may be represented as "Machine Learning", but a patent document written in English may be represented as "Machine Learning", so that the server 10 may be complex in both cases. Acquire a noun phrase set to increase the efficiency of the search.

Meanwhile, in the above-described embodiment, the condition in which the preset code includes the parentheses is described, but the present invention is not limited thereto. According to various embodiments of the present disclosure, the predetermined sign may be a sign such as a hyphen (-), a slash (/), a tilde (~), a quote ('or "), or the like.

Thereafter, the server 10 may determine the changed plurality of words as candidates for the compound noun phrase set.

Meanwhile, according to various embodiments of the present disclosure, when the sentence included in the target patent document has a preset structure, the server 10 may include at least one word except for a word corresponding to the preset structure among words included in the sentence. Can be obtained.

For example, the predetermined structure may be a structured structure such as "to be defined as". Specifically, in the target patent document, "to be defined as", "" hereinafter is referred to as. " "Can be understood as." A sentence of structure such as "describes" may be used to define terms that are not commonly used in the art. Accordingly, the server 10 may determine a sentence defining a term not generally used in the technical field, and obtain a word defined in the determined sentence.

For example, if the sentence to be analyzed is a sentence "Machine learning is described (or defined) as OOO", the server 10 determines that the sentence is a predetermined structure, and the machine, running, OOO Words can be obtained.

Thereafter, the server 10 may obtain semantic information of at least one word.

In detail, when the acquired word of OOO has semantic information commonly used in the technical field, the server 10 may match semantic information corresponding to OOO with OOO. However, when the term OOO is newly defined by the user and the semantic information is not clear, the server 10 may match the semantic information corresponding to the machine learning with the semantic information of the OOO.

Thereafter, the server 10 may obtain a word corresponding to the semantic information.

However, the present disclosure is not limited thereto, and the server 10 may replace the word of OOO with a word of machine learning when the degree of association between machine learning and OOO is equal to or greater than a preset degree of association.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may acquire a word except for a template of a patent document.

In detail, the patent document is a formal document, which is not significantly related to the essential contents of the present invention, but there may exist a paragraph describing a necessary essential component in a formal method. Therefore, when the formal paragraph is removed and a clue set is obtained, there is an effect of reducing the amount of computation of the server 10.

Accordingly, the server 10 may obtain a plurality of identification item information and a template included in the target patent document.

In one embodiment, the template may be set to a paragraph existing at a specific location. For example, the template may include the word of FIG. 1 for the first time in the [specific information for carrying out the invention] identification item included in the sentence following the [specific information for carrying out the invention] identification item included in the target patent document. May be the text up until the sentence being searched for.

In another embodiment, the template may mean a part in common with other patent documents prepared by the creator of the target patent document to be analyzed. That is, the server 10 determines the creator of the target patent document, obtains another patent document of the determined author, compares the obtained plurality of patent documents, and judges that the similar ratio is equal to or more than a preset ratio through the comparison result. Paragraphs or identification items can be obtained as templates. Since the patent document created by the same author is likely to have a similar sentence structure, the criteria of the server 10 similar judgment may be set to a paragraph or an identification item instead of a sentence unit.

In another embodiment, the server 10 may obtain a text corresponding to a predetermined identification item as a template. For example, the server 10 may obtain the text included in the identification item such as [a brief description of the drawings], [description of the reference] as a template.

In the above-described embodiment, a method for obtaining a template for a general identification item of a patent document for a domestic application has been described, but is not limited thereto. That is, the server 10 may identify various identification items, such as identification items of the national document for PCT applications, identification items of English forms for PCT applications, and furthermore, identification items of forms used in US, Japan, China, and European applications. Of course, the template can be obtained.

Thereafter, the server 10 may obtain a plurality of sentences except for the obtained template.

Thereafter, the server 10 may assign a weight to each of the plurality of identification items.

In one embodiment, if the target patent document has the same form and identification as the present specification, the server 10 is the name of the invention, the technical field, the background of the invention, the problem to be solved, the means for solving the problem, The effects of the invention, a brief description of the drawings, specific details for carrying out the invention, a description of the signs, claims, summary, representation, and weights for each of the drawings identification items can be given.

In this case, the weight may be given by various methods. In one embodiment, the server 10 is a technical field, the background technology of the invention, claims, the effects of the invention, the specific contents for carrying out the invention, the description of the symbols, a brief description of the drawings, a summary, the problem to be solved The weights can be given in the order of the solution, the solution to the problem, and the representation.

In another embodiment, the weight is obtained for the importance score of each word (or compound noun phrase set) included in the target patent document, and based on this, the importance score for each identification item is obtained, and the weight is high for the identification item having a high importance score. Can be weighted. The importance score is an index indicating the importance of the word (or compound noun phrase set) in the patent document. A word having a high importance score (or compound noun phrase set) may be a keyword of the patent document.

Meanwhile, the importance score for each identification item may mean a value obtained by summing the importance scores of the words (or compound noun phrase sets) included in the identification item and dividing the number of words (or compound noun phrase sets). At this time, the word (or compound noun phrase set) used to obtain the importance score for each identification item may be all words (or compound noun phrase set) included in the identification item, but is not limited thereto. For example, a word (or compound noun phrase set) used to obtain an importance score for each item of identification is a word (or compound noun phrase set) that is determined by a template among all words (or compound noun phrase sets) included in the item. Of course, it can be a word (or a compound noun set) except for.

Thereafter, the server 10 may determine the priority of the identification item to perform the morphological analysis based on the weight of each piece of identification information.

Thereafter, the server 10 may obtain words for a plurality of sentences in which the template is excluded according to the determined priority.

That is, the server 10 may reduce the amount of calculation by first analyzing an identification item having a high possibility of a keyword and analyzing the identification item having a relatively low importance based on the analyzed content. Furthermore, when different semantic information about the same word is matched, the server 10 may determine the semantic information obtained from the high priority item as correct semantic information.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may acquire a word from the image even when the image is included in the target patent document.

In detail, when the image is included in the target patent document, the server 10 may determine whether an identification item adjacent to the image exists.

In one embodiment, when an identification item of [mathematical formula] exists in the target patent document and an image exists below the identification item, the server 10 may determine the image as an image for the equation. In the case of the equation, since it is likely to be the essential content of the target patent document, the server 10 may obtain the equation text included in the image.

In another embodiment, when an identification item of [Formula] exists in the target patent document and an image exists at the bottom of the identification item, the server 10 may determine the image as an image for the chemical formula.

Thereafter, when the identification item is adjacent to the image, the server 10 may analyze the image to obtain text included in the image. In this case, image analysis may be performed by optical character recognition (OCR), but is not limited thereto. For example, when the image is a chemical formula image, the server 10 may analyze the chemical formula structure to obtain a chemical formula corresponding to the image.

Thereafter, the server 10 may obtain a word based on the obtained text.

7 is a flowchart illustrating a method of calculating importance scores according to an embodiment of the present invention in detail.

In step S610, the server 10, the word appears in the first appearance ratio of the number of appearances of the word in the entire patent document to the total number of words of the patent document and the sentences of the entire patent document relative to the total number of sentences of the patent document The first detailed importance may be calculated based on the second appearance ratio of the number of appearance sentences.

Specifically, the server 10 may count the total number of words in the entire patent document, and count the number of occurrences of words in the entire patent document.

Subsequently, the server 10 may calculate the number of appearances of the words in the entire patent document as the first appearance ratio relative to the total number of words in the entire patent document.

Subsequently, the server 10 may count the total number of sentences of all patent documents, and count the number of appearance sentences in which words appear in the sentences of all patent documents.

The server 10 may calculate the total number of sentences of the entire patent document as the second appearance ratio relative to the number of sentences in which words appear in the sentences of the entire patent document.

To this end, the server 10 may search for words from the entire patent document, and search for sentences containing the words. In addition, the server 10 may analyze the sentence components of the entire patent document to count the total number of words and the total number of sentences.

Finally, the server 10 may calculate the first detail importance based on the first appearance rate and the second appearance rate.

In this case, the server 10 may calculate the first detailed importance level using Equation 1 below.

 <Equation 1>

Here, W1 is the first detail importance, wpw is the number of occurrences of words in all patent documents, WPW is the total number of words in all patent documents, wps is the number of appearance sentences in which words appear in the sentences of all patent documents. , WPS is the total number of sentences in the entire patent document, and a1 is the control constant of the second appearance ratio.

Referring to Equation 1, the server 10 may calculate the first detail importance as the number of occurrences of words in the entire patent document and the number of appearance sentences in which the words appear in the sentences of the entire patent document are smaller.

That is, the server 10 may calculate the first detailed importance as the word is duplicated in one sentence in the entire patent document.

On the other hand, the server 10 may increase the adjustment rate of the second appearance rate to increase the second appearance rate even if the number of appearance sentences with words appear in the sentences of the entire patent document, and decrease the adjustment constant of the second appearance rate As a result, even if the number of sentences in which words appear in the sentences of the entire patent document is high, the second appearance ratio can be reduced.

In an embodiment, it is noted that the type of the method of calculating the importance of detail is not limited as long as the server 10 calculates the first detail importance of the word in the target patent document.

For example, the server 10 may calculate the first detail importance using a TF-IDF (Term Frequency-Inverse Document Frequency) analysis method that calculates importance based on the appearance frequency as one of the text analysis methods.

In step S620, the server 10, the word appears in the sentence of the entire patent document relative to the total number of words of the patent classification information, the third appearance ratio of the number of appearance of the word in the patent classification information and the total number of sentences of the patent document The second detailed importance may be calculated based on the fourth appearance ratio of the number of appearance sentences.

In detail, the server 10 may count the total number of words of the patent classification information and count the number of occurrences of words in the patent classification information.

Here, the patent classification information is a code that can classify patents according to the technical field, and may be any one of International Patent Classfication (IPC), Cooperative Patent Classification (CPC), and F-Term.

Thereafter, the server 10 may calculate the number of appearances of the words in the patent classification information as the third appearance ratio to the total number of words in the patent classification information.

Subsequently, the server 10 may count the total number of sentences of the patent classification information, and count the number of appearance sentences in which words appear in the sentences of all patent documents.

The server 10 may calculate the total number of sentences of the entire patent document as the fourth appearance ratio from the number of appearance sentences in which words appear in the sentences of the entire patent document.

To this end, the server 10 may search for words from the patent classification information and search for sentences containing the words. In addition, the server 10 may analyze the sentence components of the patent classification information to count the total number of words and the total number of sentences.

Finally, the server 10 may calculate the second detailed importance based on the third appearance rate and the fourth appearance rate.

At this time, the server 10 may calculate the second detailed importance level using Equation 2 below.

 <Equation 2>

Where W ₂ is the second level of importance, ipcw is the number of occurrences of words in the patent classification information, IPCW is the total number of words in the patent classification information, and ipcs is the number of appearance sentences in which words appear in the sentences of the entire patent document. Where IPCS is the total number of sentences in all patent documents, and a2 is the adjustment constant of the fourth occurrence rate.

Referring to Equation 2, as the number of occurrences of words in the patent classification information and the number of appearance sentences in which words appear in the sentences of the entire patent document are smaller, the second detailed importance may be calculated.

That is, the server 10 may calculate the second detailed importance as the word is duplicated in one sentence in the patent classification information.

On the other hand, the server 10 may increase the fourth appearance rate by increasing the adjustment constant of the fourth appearance rate, even if the number of sentences in which the word appears in the sentence of the entire patent document, and decrease the adjustment constant of the fourth appearance rate In this way, even if the number of appearance sentences in which words appear in the sentences of the entire patent document can reduce the fourth appearance ratio.

In an embodiment, as long as the server 10 calculates the second detailed importance of the word in the patent document, it is noted that the kind of the detailed importance calculation method is not limited.

For example, the server 10 may calculate the second detailed importance using a term frequency-inverse document frequency (TF-IDF) analysis that calculates importance based on the appearance frequency as one of the text analysis methods.

In operation S630, the server 10 may calculate an influence value of each of the search patent documents based on reference information of each of the search patent documents, and calculate a third detailed importance of the search patent document using the influence value.

In detail, the server 10 may search for a search patent document including a word that is an object of calculation of an importance score among all patent documents.

The server 10 may calculate an influence value of each search patent document based on reference information of each search patent document.

Specifically, the server 10 may calculate the influence value based on the number of items of the applicant, inventor, and right holder, which are the reference information of the search patent document, as the same as the other patent document, the number of citations and the number of citations which are reference information. have.

That is, the server 10 may calculate the degree to which the search patent document is related to another patent document as an influence value. For example, when the search patent document is cited from several patent documents, the server 10 may determine that the search patent document has an influence on the patent document and calculate the search patent document as an influence value of the search patent document.

The server 10 may calculate the third detailed importance of the search patent document by using the influence value calculated for each search patent document.

In detail, the server 10 may calculate an average of the influence values calculated for each of the search patent documents, and calculate the calculated average as the third detailed importance.

At this time, the server 10 compares each of the minimum detail importance value and the first detail importance, the second detail importance, and the third detail importance with respect to each of the first detail importance, the second detail importance, and the third detail importance. In this case, the recalculation process may be performed for the subtle importance level below the minimum detail importance value.

Thereafter, when the first detail importance level, the second detail importance level, and the third detail importance level are greater than or equal to the minimum detail importance value, the server 10 adds a plurality of the first detail importance level, the second detail importance level, and the third detail importance level to an importance score. Can be calculated.

The server 10 may output the calculated importance score to an internal memory or a processor or transmit the calculated importance score to an external server.

8A and 8B are flowcharts illustrating a method for obtaining a similar patent document and a method for determining a similar sentence according to an embodiment of the present invention.

In detail, as illustrated in FIG. 8A, in operation S705, the server 10 may acquire a plurality of words included in the target patent document.

In operation S710, the server 10 may obtain a plurality of target patent clusters by clustering the obtained plurality of words.

In this case, various clustering techniques may be applied to clustering according to the present invention. In an embodiment, the server 10 may acquire a plurality of target patent clusters by using an Elasticsearch clustering clustering technique. In another embodiment, the server 10 may include a plurality of targets using at least one of K-means clustering, Density-based spatial clustering of applications with noise (DBSCAN) clustering, Hierarchical clustering, and mixed Gaussian clustering. Patent clusters can be obtained.

In operation S715, the server 10 acquires a plurality of midpoints for each of the plurality of target patent clusters, and based on the number of words included in the plurality of midpoints and the target patent cluster for each of the plurality of target patent clusters. The position of the patent document can be determined.

In one embodiment, the center point of the target patent cluster may be obtained based on the position of a plurality of words included in the cluster, and the position of the target patent document may be the center of gravity of the center points of the plurality of target patent clusters. have.

In operation S720, the server 10 may obtain a plurality of words included in a patent document and cluster the obtained plurality of words to obtain a plurality of patent clusters.

In step S725, the server 10 acquires a plurality of midpoints for each of the plurality of patent clusters, and positions the patent document based on the number of words included in the plurality of midpoints and the patent clusters for each of the plurality of patent clusters. Can be determined.

In one embodiment, the center point of the patent cluster may be obtained based on the position of a plurality of words included in the cluster, and the position of the patent document may be the center of gravity of the center points of the plurality of patent clusters.

In operation S730, when the location of the target patent document and the location of the patent document are within a predetermined distance, the server 10 may determine the patent document as a similar patent document.

In one embodiment, the server 10 may determine a plurality of patent documents existing within a predetermined distance from the position of the target patent document as the similar patent document, the plurality of determined in the order of similar patent documents close to the target patent document You can sort similar patent documents.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may obtain a plurality of sentences from the target patent document and the similar patent document, and sort them in order of importance.

Specifically, the server 10 may obtain a plurality of words for each of a plurality of sentences included in the target patent document and the similar patent document.

Thereafter, the server 10 may obtain a plurality of importance scores for each of the obtained plurality of words. The method of obtaining the importance score will be described later.

Thereafter, the server 10 may determine the importance score of the sentence included in the target patent document based on the plurality of importance scores.

In an embodiment, the server 10 may obtain a degree of association between the plurality of words and the target patent document, and determine the sum of the importance scores of the words having the obtained degree of association greater than or equal to a preset value as the importance score of the sentence. . In this case, the degree of association may be obtained through various methods. For example, the degree of association between the word and the target patent document may be obtained based on a plurality of patent documents corresponding to the IPC field corresponding to the target patent document. Specifically, the degree of association between a word and a target patent document may be obtained based on the number of times the word appears in a plurality of patent documents corresponding to the IPC field. In one embodiment, the higher the number of times a word appears in a plurality of patent documents corresponding to the IPC field, the higher the degree of association between the word and the target patent document.

Thereafter, the server 10 may sort the plurality of sentences in order of importance based on the importance score of each of the plurality of sentences.

Furthermore, the server 10 may group a plurality of sentences into n groups. In this case, a method of grouping a plurality of sentences may vary. In one embodiment, n groups may be grouped to include the same number of sentences. In another embodiment, n groups may be grouped according to a predetermined importance score. For convenience of explanation, hereinafter, it is assumed that the importance score of the sentences included in the k-th group is greater than the importance score of the sentences included in the k-th group.

The server 10 may obtain the first sentence of the target patent document and the second sentence of the similar patent document based on the importance score. In detail, the server 10 may classify the plurality of sentences included in the target patent document into n groups based on importance scores.

Thereafter, the server 10 may classify the plurality of sentences included in the similar patent document into n groups based on the importance score.

Thereafter, when the first sentence is included in the first group of the n groups of the target patent document, the server 10 may obtain the second sentence as the entire sentence included in the similar patent document.

Subsequently, when the first sentence is included in the kth group among the n groups of the target patent document, the server 10 may include the second sentence from the first group to the first group among the n groups of the similar patent document. Can be obtained from a sentence included in the (n-k + 1) group. In this case, n may be a natural number of 1 or more, and k may be a natural number of n or less.

For example, it may be assumed that the target patent document and the similar patent document are classified into three groups. In this case, when the first sentence is included in the first group, the server 10 compares the first sentence and all sentences included in the first group to the third group of the similar patent document (that is, the entire similar patent document). Evaluation results can be obtained. In addition, when the first sentence is included in the second group, the server 10 may compare the first sentence with all sentences included in the first group and the second group of the similar patent document to obtain an evaluation result. In addition, when the first sentence is included in the third group, the server 10 may compare the first sentence with all sentences included in the first group of the similar patent document to obtain an evaluation result.

That is, the server 10 obtains an evaluation result only for sentences that are considered to be important sentences according to the importance of the sentence, thereby reducing the calculation amount by preventing the evaluation result obtaining process for unnecessary or less important sentences. Specifically, since the first group of the target patent document is a high priority sentence, it is compared with the whole similar patent document, and since the third group of the target patent document is a low importance sentence, the sentence of the first group of high similarity patent documents By comparing, the server 10 can omit the comparison between the third group of (less important) target patent documents and the third group of (less important) similar patent documents.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may determine whether to group the target patent document and the similar patent document into several groups. Specifically, the server 10 may determine the value of n based on the GPU occupancy and required accuracy of the current server 10. If the GPU occupancy is high or the required accuracy is low, the server 10 may increase the value of n. If the GPU occupancy is low or the required accuracy is high, the server 10 may decrease the value of n. For example, it may be assumed that the target patent document and the similar patent document are composed of six sentences, where n is 3 and n is 6. If n is 6, server 10 compares 6 + 5 + 4 + 3 + 2 + 1 = 21 sentences; if n is 3, server 10 is 6 + 6 + 4 + 4 + 2 + 2 = 24 sentences will be compared. That is, as the value of n increases, the number of sentences to be compared in the target patent document and the similar patent document decreases. As the value of n decreases, the number of sentences to be compared in the target patent document and the similar patent document increases, so that the server 10 If the amount of computation (GPU occupancy) that is currently handled by him or the required accuracy is low, the server 10 may increase the value of n.

Meanwhile, in the above-described embodiment, the case where the target patent document and the similar patent document are grouped by the same n number has been described, but is not limited thereto. For example, the target patent document and the similar patent document can be grouped in different numbers.

Meanwhile, as shown in FIG. 8B, in step S735, the server 10 may obtain similarity scores and dissimilarity scores of the first sentence and the second sentence, respectively.

In detail, the similarity score is an index for determining whether the first sentence and the second sentence are related sentences, and the dissimilarity score is an index for determining whether the first sentence and the second sentence are irrelevant sentences.

In one embodiment, the similarity score and the dissimilarity score may be obtained by obtaining semantic information of a word included in the first sentence and the second sentence, and determining whether the obtained semantic information is semantic information on the same subject. have.

In another embodiment, the server 10 may determine the sentence structure of the first sentence and the second sentence and use it to calculate the similarity score and dissimilarity score. For example, the server 10 may increase the similarity score when the sentence structures of the first sentence and the second sentence are similar, and increase the dissimilarity score when the sentence structures of the first sentence and the second sentence are dissimilar. .

In operation S740, when the dissimilarity score is greater than or equal to a predetermined score, the server 10 may determine that the first sentence and the second sentence are irrelevant sentences.

In operation S745, when the similarity score is equal to or greater than a predetermined score, the server 10 may obtain at least one word not included in the second sentence among words included in the first sentence.

That is, even if the first sentence and the second sentence are related sentences, the first sentence and the second sentence are not sentences having the same meaning. Accordingly, the server 10 may determine whether the first sentence and the second sentence are the same sentence or a sentence about the same subject or a difference exists.

In an embodiment, the server 10 may determine at least one word not included in the second sentence among words included in the first sentence in order to determine a difference between the first sentence and the second sentence.

In another embodiment, the server 10 may determine the sentence structure of the first sentence and the second sentence.

In operation S750, the server 10 obtains at least importance scores for each of the at least one word not included in the second sentence, and determines whether there is a word equal to or greater than a predetermined importance score among the obtained at least one importance scores. You can judge.

In operation S755, when there is no word having a predetermined importance score or more, the server 10 may determine the first sentence and the second sentence as the matching sentence.

In operation S760, when there is a word equal to or greater than a predetermined importance score, the server 10 may determine the first sentence and the second sentence as inconsistent sentences.

That is, even when at least one word of the first sentence is not included in the second sentence, a word having a low importance score may not be associated with the similarity between the first sentence and the second sentence. Therefore, the server 10 obtains a word when the importance score of at least one word not included in the second sentence is equal to or greater than a preset importance score, and the first sentence and the second sentence are matched sentences based on the acquired word. It can be determined whether or not the sentence is a mismatch.

In this case, in step S180, the server 10 may not include the second sentence in the generated prior art research report when the first sentence and the second sentence are not related. The server 10 may generate a prior art research report based on a second sentence identical to the first sentence or a second sentence similar to the first sentence.

According to an embodiment of the present disclosure, the server 10 may obtain a plurality of sentences of the target patent document and sort them in order of importance, and determine whether there are similar sentences in the similar patent document from the sentences of high importance. For example, the server 10 may sort the target patent document in order of importance, and obtain the first-first sentence to the first-n sentence.

The server 10 may obtain a second sentence of the similar patent document with respect to the first-first sentence by comparing the first-first sentence with the highest priority with the similar patent document. In the same manner, the server 10 obtains a second sentence of the similar patent document for the 1-2 sentence by comparing the 1-2 patent and the similar patent document, and compares the 1-n sentence and the similar patent document. The second sentence of the similar patent document with respect to the first-n sentence can be obtained.

The server 10 may select only a sentence related to the first sentence among the obtained plurality of second sentences. According to various embodiments of the present disclosure, the server 10 may display second sentences for all sentences (ie, first-first sentence to first-n sentence) of the target patent document, but is not limited thereto. Of course, only the second sentence of a predetermined upper priority sentence (for example, a sentence corresponding to the top 30% as a result of sorting in the order of importance) of the patent document may be displayed.

On the other hand, according to various embodiments of the present invention, the server 10 is included in the claims of the target patent document, at least one similar patent document for the target patent document, the target patent document and at least one similar patent document claims Based on the importance score for the word, it is possible to determine whether a specific sentence of the target patent document violates the similar patent document.

9 is a flowchart illustrating a specific method of generating a thesaurus according to an embodiment of the present invention.

In operation S810, the server 10 may convert a word into a word vector through Word2Vec learning based on the patent document.

In operation S820, the server 10 may calculate a distance between any two word vectors among the word vectors, and calculate the calculated distance as the similarity.

In operation S830, the server 10 determines whether the similarity between any two word vectors of the word vectors is less than the preset reference similarity, and if the similarity between any two word vectors of the word vectors is less than the preset reference similarity, the corresponding two words. Two words corresponding to the vector may be grouped into a group of similar words.

In operation S840, the server 10 may determine whether any one of the acquired patent documents satisfies the noise document condition, and may remove the patent document that satisfies the noise document condition.

Meanwhile, according to various embodiments of the present disclosure, the server 10 may select a reference word vector from among word vectors.

Here, the reference word vector may be a word vector corresponding to the reference word of the similar word group described later. For example, if the reference word is an apple, words included in the similar word of the apple may be included in the similar word group.

Thereafter, the server 10 may calculate the similarity between the reference word vector and another word vector. In this case, the server 10 may calculate the similarity between the reference word vector and another word vector by the same method as the above-described similarity calculation method.

Subsequently, the server 10 may determine whether a word corresponding to the other word vector is included in the similarity group based on the reference word corresponding to the reference word vector, based on the similarity between the reference word vector and the other word vector.

Specifically, if the similarity between the reference word vector and the other word vector is less than the preset reference similarity, the server 10 includes the word corresponding to the other word vector in the similarity group based on the reference word corresponding to the reference word vector. You can check.

Subsequently, when the server 10 determines that the word corresponding to the other word vector is included in the similarity group based on the reference word corresponding to the reference word vector, the server 10 determines the similarity based on the reference word based on the word corresponding to the other word vector. Can be included in a group.

10 is an exemplary diagram for describing a process of deriving an evaluation result using the first AI model and the second AI model according to an embodiment of the present invention.

As illustrated in FIG. 10, the first artificial intelligence model may receive a target patent document as an input value and obtain a similar patent document for the target patent document. In this case, as well as the number of similar patent documents for the target patent document, of course, if there are a plurality of similar patent documents, the server 10 may arrange the plurality of similar patent documents based on the association with the target patent document. Of course.

The first database is a configuration for storing the entire patent document. Specifically, the first database obtains words of each patent document through morphological analysis of each patent document, obtains a thesaurus of the acquired words, and obtains an importance score for each of the acquired words. The acquired importance score may be matched with a word and stored.

The first artificial intelligence model may obtain a similar patent document corresponding to the target patent document among all patent documents stored in the first database.

Meanwhile, when the target patent document and the similar patent document are obtained through the first artificial intelligence model, the server 10 may extract the first sentence of the target patent document and extract the second sentence of the similar patent document. The first sentence and the second sentence may be obtained through the various methods described above.

In addition, the second AI model may obtain the evaluation result of the first sentence and the second sentence by inputting the first sentence and the second sentence as input values. In this case, the evaluation result may be a result of whether the first sentence and the second sentence are matched sentences, mismatched sentences, or irrelevant sentences, and may be obtained according to the above-described various embodiments.

The second database may store a patent document and a previous search report matching the patent document. In this case, the preceding research report may include information on a specific patent of a similar patent document corresponding to a specific sentence of a similar patent document and a patent document determined to be similar to the patent document.

11 is a block diagram of an apparatus according to an embodiment of the present invention.

The processor 102 may include a connection passage (eg, a bus, etc.) that transmits and receives signals with one or more cores (not shown) and graphics processing unit (not shown) and / or other components. .

The processor 102 according to one embodiment executes one or more instructions stored in the memory 104 to perform the method described in connection with FIGS. 1-9.

For example, the processor 102 obtains at least one word based on the target patent document by executing one or more instructions stored in memory, obtains an importance score of the obtained at least one word, and retrieves the target patent document. 1 inputting an artificial intelligence model to obtain a similar patent document, to obtain a plurality of sentences contained in the target patent document, to obtain a plurality of sentences included in the similar patent document, and among the plurality of sentences included in the target patent document Acquire a first sentence, obtain a second sentence among a plurality of sentences included in the similar patent document, input the first sentence and the second sentence into the second artificial intelligence model, and Obtaining the evaluation result, based on the evaluation result, it is possible to generate a prior art research report on the target patent document.

Meanwhile, the processor 102 may include random access memory (RAM) and read-only memory (ROM) for temporarily and / or permanently storing a signal (or data) processed in the processor 102. , Not shown) may be further included. In addition, the processor 102 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processor, a RAM, and a ROM.

The memory 104 may store programs (one or more instructions) for processing and controlling the processor 102. Programs stored in the memory 104 may be divided into a plurality of modules according to functions.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, in a software module executed by hardware, or by a combination thereof. The software module may be a random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art. The components of the present invention may be embodied as a program (or an application) and stored in a medium for execution in combination with a computer which is hardware. The components of the present invention may be implemented in software programming or software elements, and similarly, embodiments include C, C ++, including various algorithms implemented in combinations of data structures, processes, routines, or other programming constructs. It may be implemented in a programming or scripting language such as Java, assembler, or the like. Functional aspects may be implemented in algorithms running on one or more processors.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains may realize the present invention in other specific forms without changing the technical spirit or essential features thereof. I can understand that. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive.

10: server
20: user terminal

Claims (10)

In the control method of the similarity determination system of a patent document,
Acquiring, by the server, a target patent document;
Acquiring, by the server, at least one word based on the target patent document;
Acquiring, by the server, an importance score of the obtained at least one word;
Obtaining, by the server, a similar patent document by inputting the target patent document into a first artificial intelligence model;
Acquiring, by the server, a plurality of sentences included in the target patent document, and obtaining a plurality of sentences included in the similar patent document;
Acquiring, by the server, a first sentence among a plurality of sentences included in the target patent document and a second sentence among a plurality of sentences included in the similar patent document;
Acquiring, by the server, the first sentence and the second sentence into a second artificial intelligence model to obtain evaluation results of the first sentence and the second sentence; And
Generating, by the server, a prior art research report on the target patent document based on the evaluation result;
Acquiring the word,
Acquiring, by the server, a morpheme based on the target patent document to obtain a plurality of words;
Determining, by the server, an error word among the obtained plurality of words;
The server correcting the error word; And
And obtaining, by the server, at least one compound noun phrase set based on the degree of association between the plurality of words. delete Claim 3 has been abandoned upon payment of a setup registration fee. The method of claim 1,
Acquiring the importance score,
Acquiring, by the server, a word that is an object of calculating an importance score from the target patent document;
The server includes the first detail importance of the word in the entire patent document, the second detail importance of the word in the patent classification information corresponding to the technical field information of the target patent document, and the word among the entire patent document. Calculating at least one detail importance of the third detail importance of the searched patent document; And
And calculating, by the server, the importance score of the word based on at least one of the first detail importance level, the second detail importance level, and the third detail importance level. The method of claim 1,
The control method,
Acquiring, by the server, an entire patent document that is the basis of a thesaurus;
Analyzing, by the server, morphemes for each of the entire patent documents;
Converting a word included in each of the entire patent documents into a word vector based on the result of the morphological analysis;
Calculating, by the server, the similarity between the word vectors; And
And the server grouping a word corresponding to the word vector into a group of similar words based on the similarity. The method of claim 1,
Determining the error word,
Matching semantic information with respect to each of the obtained plurality of words;
Obtaining an accuracy score for each of the plurality of words based on the matched semantic information and the target patent document; And
Determining a word having an accuracy score equal to or less than a preset value as an error word,
Correcting the error word,
Obtaining a plurality of semantic information on the error word;
Obtaining a plurality of weights for each of the obtained plurality of semantic informations;
Obtaining a weight having the highest association with the target patent document among the plurality of weights; And
Matching semantic information corresponding to the weight having the highest association with the target patent document to the error word;
Obtaining the at least one compound noun phrase set,
Obtaining a plurality of words for sentences included in the target patent document;
Obtaining a plurality of compound noun phrase set candidates obtained by combining the plurality of words;
Obtaining a frequency at which a compound noun phrase set identical to the obtained plurality of compound noun phrase set candidates is included in the target patent document; And
Determining a compound noun phrase set candidate having the frequency equal to or greater than a preset frequency as a compound noun phrase set; Including,
And the compound noun phrase set candidate comprises sequence information and separation information of words included in the compound noun phrase set candidate. The method of claim 3,
The step of calculating the detailed importance of the word
The first occurrence rate of the number of occurrences of the word in the entire patent document relative to the total number of words in the entire patent document and the number of appearance sentences in which the word appears in the sentences of the entire patent document relative to the total number of sentences in the total patent document Calculating the first sub-critical importance based on a second occurrence ratio of;
The third occurrence rate of the number of occurrences of the word in the patent classification information to the total number of words of the patent classification information and the number of appearance sentences in which the word appeared among the sentences of the patent document compared to the total number of sentences of the patent document Calculating the second sub-critical importance based on a fourth appearance ratio of the second; And
Calculating an influence value of each of the search patent documents based on reference information of each of the search patent documents, and calculating a third detailed importance of the search patent document using the influence value;
The calculating of the first detail importance level
The first detailed importance is calculated using Equation 1 below.
The calculating of the second detailed importance level
A method of controlling a patent document similarity determination system comprising calculating the second detailed importance level using the following equation.
<Equation 1>

<Equation 2>

Here, the W ₁ is the first detail importance, the wpw is the number of occurrences of the word in the entire patent document, the WPW is the total number of words in the patent document, the wps is among the sentences of the patent document The word is the number of appearance sentences appeared, the WPS is the total number of sentences of the entire patent document, the a1 is a control constant of the second appearance ratio, the W ₂ is a second detail importance, the ipcw is the The number of occurrences of the word in patent classification information, the IPCW is the total number of words in the patent classification information, the ipcs is the number of appearance sentences where the word appeared in the sentences of the entire patent document, the IPCS is the total patent A total is the number of sentences in the document, and a2 is a control constant of the fourth appearance ratio. Claim 7 was abandoned upon payment of a set-up fee. The method of claim 1,
Acquiring the similar patent document,
Acquiring a plurality of words included in the target patent document;
Clustering the obtained plurality of words to obtain a plurality of target patent clusters;
Acquiring a plurality of midpoints for each of the plurality of target patent clusters, and determining a position of the target patent document based on the plurality of midpoints for each of the plurality of target patent clusters and the number of words included in the target patent cluster. Making;
Obtaining a plurality of words included in a patent document and clustering the obtained plurality of words to obtain a plurality of patent clusters;
Acquiring a plurality of midpoints for each of the plurality of patent clusters, and determining a position of the patent document based on the plurality of midpoints for each of the plurality of patent clusters and the number of words included in the patent cluster; And
And determining the patent document as the similar patent document when the position of the target patent document and the position of the patent document are within a preset distance.
Obtaining an evaluation result for the first sentence and the second sentence,
Obtaining similarity scores and dissimilarity scores of the first sentence and the second sentence, respectively;
Determining that the first sentence and the second sentence are irrelevant sentences when the dissimilarity score is equal to or greater than a predetermined score;
Obtaining at least one word not included in the second sentence among words included in the first sentence when the similarity score is equal to or greater than a predetermined score;
Obtaining at least importance scores for each of at least one word not included in the second sentence, and determining whether there is a word equal to or greater than a predetermined importance score among the obtained at least one importance scores;
Determining that the first sentence and the second sentence are matched sentences when a word having a predetermined importance score or more does not exist; And
And determining the first sentence and the second sentence as inconsistent sentences when a word having a predetermined importance score or more exists. The method of claim 4, wherein
Converting to the word vector
Converting the word into a word vector through Word2Vec learning based on the patent document;
Computing the similarity
Calculating a distance between any two word vectors of the word vectors and calculating the calculated distances with similarities;
Grouping into the group of similar words
Determine whether the similarity between any two word vectors of the word vectors is less than a preset reference similarity, and if the similarity between any two word vectors is less than a preset reference similarity, two words corresponding to the two word vectors Grouping into a group of similar words;
Acquiring the entire patent document
And determining whether any one of the acquired patent documents satisfies the noise document condition, and removing the patent document that satisfies the noise document condition. Memory for storing one or more instructions; And
A processor for executing the one or more instructions stored in the memory;
The processor executes the one or more instructions,
An apparatus for carrying out the method of claim 1. A computer program, coupled with a computer, which is hardware, stored on a recording medium readable by a computer to perform the method of claim 1.

Images