JP2003281138A - System and method of processing natural language, and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately output the semantic information of a adverbial particle which may have a plurality of semantic functions according to the position of appearance. <P>SOLUTION: The adverbial particle 'bakari' has two meanings: a meaning of 'limitation' and a meaning of 'degree'. The information about the sentence structure of the adverbial particle 'bakari' included in an inputted sentence is extracted, and whether the word 'bakari' relates to a quantifier or not is determined. When the result of determination is positive, the meaning of 'bakari' is decided to mean the 'degree', and to mean the 'limitation' in other cases, and the result of the semantic analysis is outputted. For example, in the sentence 'Kare ha ringo wo 3-ko bakari tabeta' (He ate some three apples), the meaning of 'bakari' is decided to mean the 'degree' because this 'bakari' relates to the quantifier '3-ko'. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a natural language processing system and a natural language processing method for mathematically handling a natural language used by humans for daily communication.
The present invention also relates to a computer program, and more particularly to a natural language processing system and a natural language processing method for performing syntactic and semantic analysis of Japanese syntax, and a computer program.

More specifically, the present invention relates to a natural language processing system and a natural language processing method for performing syntactic / semantic analysis of Japanese syntax according to grammatical rules, and a computer program. The present invention relates to a natural language processing system, a natural language processing method, and a computer program that accurately output the semantic information of a part of speech having the meaning of.

[0003]

2. Description of the Related Art Words used by humans for daily communication such as Japanese and English are called "natural language". Many natural languages have a naturally occurring origin and have evolved with the history of humanity, ethnicity, and society. Of course, a person can communicate by gesturing or gesturing, but natural language can realize the most natural and advanced communication.

On the other hand, with the development of information technology, computers have become established in human society and have been deeply permeated in various industries and daily lives. Now, not only computer data but almost all information contents such as images and sounds are handled on the computer, and it is possible to perform advanced processing such as editing / processing, storage, management, transmission and sharing of information. .

Natural language has an abstract and ambiguous nature by nature, but by treating sentences mathematically,
Computer processing can be performed. As a result, various applications related to natural language such as machine translation, dialogue system, search system, etc.
The service is realized.

Natural language processing is generally divided into processing phases of morphological analysis, syntactic analysis, semantic analysis, and context analysis.

In the morphological analysis, a sentence is segmented into morphemes, which are the smallest semantic units, and a part-of-speech recognition process is performed. In the syntactic analysis, a sentence structure such as a phrase structure is analyzed based on grammatical rules. Since the grammar rules have a tree structure,
The syntactic analysis result generally has a tree structure in which individual morphemes are joined based on the dependency relationship. In the semantic analysis, based on the meaning (concept) of the words in the sentence and the semantic relationship between the words, the semantic structure expressing the meaning conveyed by the sentence is obtained, and the semantic structures are synthesized. In context analysis, a sentence (discourse), which is a series of sentences, is regarded as a basic unit of analysis, and a discourse structure is constructed by obtaining a semantic unity between sentences.

By the way, among the many natural languages, Japanese is said to be highly ambiguous. One of the reasons depends on the existence of a part-of-speech, which is generally called a sub particle.

For example, "do", "just", "only"
Adverbs such as, etc. have the property of appearing at various positions in the sentence (can be attached after any of nouns, verbs, particles, and adjectives), and one word may have multiple semantic functions. It has the characteristic of being. That is, since the meaning of the sub particle changes depending on the usage, it increases the ambiguity of the sentence and makes it difficult to correctly interpret and translate the sentence.

For example, the following two sentence examples both include the sub particle "saku".

(A) He eats only apples. (B) He ate only three apples.

In the sentence example (a), the auxiliary particle "saku" appears at the position where the case particle "wo" indicating the object of the verb is located.
On the other hand, in the sentence example (b), the subposition particle “saku” is the quantifier “3
It appears as a suffix of "individual". And, “saku” in sentence example (a) means “limited”, which means “eating more apples than other things”, but “saku” in sentence example (b) is “3”. It means the "degree" of "about individual pieces".

When translating each of the above sentence examples from Japanese to English, it is indispensable to know what meaning the sub particle is used for. Because the translation of sentence example (a) is "He eats apples only", but the sentence example (b) is "He eats apples only"
because it must be translated as "ate about three apples".

In the conventional syntax analysis system, etc., the sub particle "simple" used in the usage as shown in the sentence example (a).
Is a substitute for the "wo" case, which indicates the object, and the auxiliary particle "sword" in the usage shown in sentence example (b) is equivalent to the suffix,
The syntactic information is output.

However, the above-mentioned "limitation",
Since the semantic information such as “degree” is not output, it is difficult to eliminate the ambiguity when the sub particle has a plurality of meanings. It is an important task to accurately analyze the meaning of a sub particle because the expressions including a sub particle are used in the Japanese language very standardly.

<< Commentary >> There are four types of particles: "case particle", "connecting particle", "sub particle", and "final particle". The case particle is attached to a word such as “ga”, “de”, and “no” that is not used, such as a word, and has a function of expressing a case relationship between the word and another word. Conjunctive particles have a function to attach conditions or parallels to verbs, adjectives, and auxiliary verbs such as "ba", "domo", and "tokoroga" and connect the sentence or clause to the sentence or clause below. . The adverbial particle has a function of adding various meanings such as limitation, degree, emphasis, and analogy to a word or adverb such as "sahe" or "only". The final particle is "ka",
It is used at the end of a sentence or phrase such as "na" or "domo" to connect the sentence or clause to the sentence or clause to indicate the attitude or feeling of the speaker such as conditions or parallelism. The particle is "ha",
For words related to predicates such as "mo" and "shika", it has an effect on the statement of the word, and at the end of the phrase, it has the function of helping the formation of a sentence. The interjection particle is "ya",
It is inserted between breaks such as "o" and "shi", and has the function of adding vocabulary, adjusting tone, and adding feelings.

[0017]

SUMMARY OF THE INVENTION An object of the present invention is to provide an excellent natural language processing system, natural language processing method, and computer program capable of performing syntactic and semantic analysis of Japanese syntax according to grammatical rules. To do.

A further object of the present invention is to accurately determine the semantic information of a part of speech having a plurality of semantic functions.
An object is to provide an excellent natural language processing system, natural language processing method, and computer program.

A further object of the present invention is to provide an excellent natural language processing system, natural language processing method, and computer program capable of accurately outputting the semantic information of a part of speech having a plurality of meanings depending on the method of use. To provide.

A further object of the present invention is to provide an excellent natural language processing system and natural language processing method capable of accurately outputting the semantic information of a subparticle that may have a plurality of semantic functions depending on the appearance position. , And providing a computer program.

[0021]

The present invention has been made in consideration of the above problems, and a first aspect thereof is a natural language system for analyzing semantic information of a part of speech having a plurality of meanings, or A natural language processing method, which is a meaning candidate setting means or step for setting meaning candidates in a part of speech having a plurality of meanings included in an input sentence, and a syntactic information extraction for extracting syntactic information of the part of speech included in the input sentence. A natural language system comprising: a means or step; and a meaning output means or step for selecting one of the meaning candidates as the meaning of the part of speech according to the extracted syntactic information of the part of speech. It is a natural language processing method.

The part of speech having a plurality of meanings referred to here is, for example, an auxiliary particle. Adverbs such as “ho”, “just”, and “only” have the property of appearing at various positions in a sentence, and one word may have a plurality of semantic functions. Since the meaning of the sub particle changes depending on the usage, it increases the ambiguity of the sentence and
Making translation difficult.

According to the natural language system or the natural language processing method according to the first aspect of the present invention, the meaning candidate setting means or step searches the dictionary describing the meaning function of the part of speech to find the candidate of the part of speech. get.

Then, the syntactic information extraction means extracts the related party or the connection destination as the syntactic information of the part of speech. For example, by using an input sentence that has undergone syntax analysis, cs
Since the phrase structure of sentences (such as dependency relations and connection of parts of speech) is clearly described like "structure",
It is possible to easily acquire the part-of-speech syntax information.

Then, the meaning output means or step selects one of the meaning candidates according to the extracted syntactic information of the part of speech.
Choose one as the meaning of the part of speech. As a result, the meaning in the input sentence of the part of speech having multiple meaning candidates is clarified,
It can be output as a semantic analysis result.

The meaning output means or step is, for example,
If the syntactic subordinate is a quantifier, the meaning of the auxiliary particle “saku”, which has two meanings, “degree” and “limit”, is determined to be “degree”, and otherwise it is determined to be “limit”. You may do it.

Further, the meaning output means or step may be used, for example, when the sub particle "up to" having two meanings of "emphasis" and "degree" is not connected to the case particle or the omitted case particle. When the omitted case particle “ni” or “he” is connected, its meaning may be determined as “degree”, and in other cases, it may be determined as “emphasized”.

It can be provided as a grammar rule description that the meaning function of a part of speech such as a subparticle having a plurality of meanings is determined according to the syntactic information of the part of speech in the sentence. In addition, by preparing as a grammar rule description, it is possible to output the correct meaning of the auxiliary particle included in the input sentence by performing a normal syntactic / semantic analysis process.

The second aspect of the present invention is a computer program written in a computer-readable format so as to execute natural language processing for analyzing semantic information of a part of speech having a plurality of meanings on a computer system. There
A meaning candidate setting step of setting a meaning candidate in a part of speech having a plurality of meanings included in the input sentence, a syntax information extraction step of extracting syntax information of the part of speech included in the input sentence, and a syntax of the extracted part of speech And a meaning output step of selecting one of the meaning candidates as the meaning of the part-of-speech according to the information, the computer program.

A computer according to the second aspect of the present invention
The program defines a computer program written in a computer-readable format so as to realize a predetermined process on a computer system. In other words, by installing the computer program according to the second aspect of the present invention in the computer system, a cooperative action is exerted on the computer system, and the natural language according to the first aspect of the present invention. It is possible to obtain the same effects as those of the processing system and the natural language processing method.

Further objects, features and advantages of the present invention are as follows.
It will be apparent from the embodiments of the present invention described later and the more detailed description based on the accompanying drawings.

[0032]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below in detail with reference to the drawings.

Natural language syntactic / semantic analysis methods can be roughly classified into methods based on statistical processing and methods based on grammar rule description. The present invention can exert a remarkable effect particularly when applied to syntactic / semantic analysis based on grammar rule description.

The natural language processing system according to the present invention can be implemented, for example, by incorporating it in the syntactic / semantic analysis processing based on the LFG (Lexical-Functional Grammar) grammar theory. In LFG, the linguistic knowledge, or grammar, of native speakers is organized as a component separate from computer processing and other non-grammatical processing parameters that affect the processing behavior of the computer. First, an overview of the natural language processing system will be briefly described. Although the present embodiment is described based on the LFG grammar theory, the present invention can be similarly applied to an analysis system having other grammar rules.

FIG. 7 schematically shows the configuration of the natural language processing system 1 based on LFG.

The morpheme analysis unit 2 has a morpheme rule 2A and a morpheme dictionary 2B relating to a specific language such as Japanese, and divides an input sentence into morphemes, which are the smallest semantic units, and performs part-of-speech recognition processing. For example, if the sentence "My daughter speaks English." Is entered, the result of morphological analysis is "I {Noun} {up} daughter {Noun} is {up} English {Noun}.
{Up} speak {Verb1} {tr} and {jp}. {pt} "is output.

The result of such morphological analysis is then input to the syntactic / semantic analysis unit 3. The syntactic and semantic analysis department
It has dictionaries such as grammar rules 3A and valence valence dictionary 3B, and analyzes the phrase structure based on grammar rules and expresses the meaning conveyed by the sentence based on the meaning of words in the sentence and the meaning relation between words. Analyze the semantic structure (The valence dictionary describes the relationship between verbs and other constituent elements in the sentence, such as the subject, and can extract the semantic relationship between predicates and related words). Then, as a result of the syntactic analysis, "c-structure (constituent structur), which represents a phrase structure of a sentence including words and morphemes as a tree structure.
e) ", and as a result of semantically and functionally analyzing the input sentence based on the case structure of the subject, object, etc., such as question sentence, past tense, polite sentence," f-structure (functional stru
cture) "is output.

In FIG. 8 and FIG. 9, c-structure and f-st obtained as a result of processing the input sentence "My daughter speaks English" by the syntactic / semantic analysis unit 1.
In each of the figures, the figures are shown.

The c-structure is a tree structure representing the structure of words and phrases in a sentence, and is defined by the syntax category. For example, a phonological interpretation for generating a phoneme string can be performed based on c-structure. On the other hand, f-structure is
It is a clear expression of grammatical functions and is composed of grammatical function names, semantic forms, and feature symbols. By referring to f-structure, the subject, the object, and the complement
t) and modifiers (adjunct) can be understood. f-structure is c-structure
It is a set of features attached to each node of ure, and is expressed in the form of an attribute-attribute value matrix as shown in FIG. In other words, the left side in [] is the feature (attribute)
And the right side is the feature value (attribute value).

Regarding the details of LFG, see R.
Paper "Lexical-Functio" co-authored by M. Kaplan and J. Bresnan
nal Grammar: A Formal System for Grammatical Repre
sentation "(The MIT Press, Cambridge (1982). Repr
inted in Formal Issues in Lexical-Functional Gramma
r, pp. 29-130.CSLI publications, Stanford Univers
ity (1995).).

There are several parts of speech in Japanese,
This leads to language ambiguity. The natural language processing system according to the present invention is provided with the dictionary describing the semantic information and the grammar rule description, so that the semantic information of the part of speech is accurately output. More specifically, by providing a dictionary describing the meaning function of the adverb and a grammar rule, the semantic information of the adverb is output in the result of the semantic analysis. Also, when a word has multiple meanings, the ambiguity is resolved.

FIG. 1 shows, in the form of a flow chart, the processing procedure of the semantic function analysis of an auxiliary particle according to an embodiment of the present invention. In the example shown in the figure, it is possible to analyze the meaning function of the auxiliary particle "sword" which can have the meanings of "degree" and "limitation". Hereinafter, description will be given according to this flowchart.

As an input sentence containing the auxiliary particle "sword",
It is assumed that "he ate only three apples" was put into the processing routine. However, it is assumed that the input sentence has already been parsed, and the phrase structure of the sentence such as c-structure (dependency relation, connection relation of part of speech, etc.)
Should be clearly stated.

First, an inquiry is made to the dictionary describing the meaning function of the auxiliary particle "sword" (step S1).

FIG. 2 exemplifies the description content of the auxiliary particle "sword" in the dictionary. As shown in the figure, "just"
Since two meanings of "limitation" and "degree" are described in, the two are acquired as candidates for the meaning of "just" (step S2). If the adjunct particle has only one meaning as a result of the inquiry to the dictionary, it is sufficient to output only the single meaning.

Next, the syntactic information of the auxiliary particle "Jun" included in the input sentence is extracted to determine whether the auxiliary particle "Jun" is associated with the quantifier (step S).
3). If the result of the judgment is true, then the auxiliary particle "saku"
Is determined to be "degree" (step S4), and in other cases, the meaning of the auxiliary particle "simply" is determined to be "limited" (step S5), and the result of the semantic analysis is output (step S6). .

In this input sentence, "saku" is the quantifier "3
Is related to "individuals". Since this condition is satisfied, the result is true (step S3), the meaning of the subposition particle "just" in this case is determined to be "degree" (step S4), and the result of the semantic analysis is output (step S6). .

FIG. 3 shows the output of the semantic analysis result for the input sentence "He ate only three apples." In the example shown in the figure, the semantic analysis result is f-struc.
It is expressed in the form of a matrix of attributes, ie, attributes-attribute values, and is composed of grammatical function names, semantic forms, and characteristic symbols. The left side in [] is the name of the feature (attribute), and the right side is the value of the feature (attribute value). It should be fully understood that the meaning of the adjunct particle "sword" is explicitly expressed as shown in the figure.

For example, by preparing as a description of the grammar rule 3A that the meaning function of the auxiliary particle "sword" is determined according to the dependency relation, the syntactic / semantic analysis unit 3 performs normal semantic analysis. It is possible to output the exact meaning of the auxiliary particle “saku” included in the input sentence.

Further, FIG. 4 shows, in the form of a flow chart, the processing procedure of the semantic function analysis of the auxiliary particle according to the embodiment of the present invention. In the example shown in the figure, it is possible to analyze the meaning function of the sub particle “up to” which can have the meanings of “emphasis” and “degree”. Hereinafter, description will be given according to this flowchart.

It is assumed that "I ate an apple till him" was input to the processing routine as an input sentence containing the sub particle "to". However, it is assumed that the input sentence has already been parsed and that the phrase structure of the sentence (such as the dependency relation and the part-of-speech connection relation) is clearly described like c-structure.

First, an inquiry is made to the dictionary describing the meaning function of the sub particle "to" (step S11).

FIG. 5 exemplifies the description content of the sub particle "up" in the dictionary. As shown in the figure, since “up to” has two meanings of “degree” and “emphasis”, these two are acquired as candidates for the meaning of “up to” (step S12). In addition, as a result of inquiry to the dictionary,
When the auxiliary particle has only one meaning, it is sufficient to output only a single meaning as it is.

Next, the sub particle "to" included in the input sentence
The syntactic information of is extracted, and it is determined whether or not the sub particle "up to" is connected to the case particle or the omitted case particle (step S13). Whether or not the case particle is omitted can be known by referring to the valence dictionary of the verb (“eat” in this input sentence). The valency dictionary describes the relationship (case relationship) between a verb and other constituent elements in the sentence such as the subject, and can extract the semantic relationship between the predicate and the related word (well known).

If the sub particle "up to" is not connected to the case particle or the omitted case particle, the meaning is determined to be "degree" (step S16), and the result of the semantic analysis is output (step S17). ).

On the other hand, when the auxiliary particle “up” is connected to the case particle or the omitted case particle, the connected case particle is “ni” or “he” in which the case particle is omitted. Determine if When the sub particle “up” is connected to “ni” or “he” in which the case particle is omitted,
Similarly, the meaning is determined to be "degree" (step S1).
6), the result of the semantic analysis is output (step S17).

In this input sentence, "up to" is connected to the omitted case particle "ga", so that the result of the decision block S13 is true and the result of the following decision step S14 is false. In this case, the meaning of the sub particle "up to" is determined to be "emphasized" (step S14), and the result of the semantic analysis is output (step S17).

When the connected case particle is not the omitted "ni" or "he", the meaning of the sub particle "up to" is determined to be "emphasized" and the result of the semantic analysis is output. (Step S17).

FIG. 6 shows the output of the semantic analysis result for the input sentence "I ate an apple till him". In the example shown in the figure, the semantic analysis result is f-structure.
That is, it is expressed in the form of an attribute-attribute value matrix, and is composed of grammatical function names, semantic forms, and characteristic symbols. The left side in [] is the name of the feature (attribute), and the right side is the value of the feature (attribute value). It should be fully understood that, as shown in the figure, the meaning of the sub particle “up” is explicitly expressed.

For example, by preparing as a description of the grammar rule 3A that the meaning function is determined according to the dependency relationship of the auxiliary particle "sword", the syntactic / semantic analysis unit 3 performs normal semantic analysis. It is possible to output the exact meaning of the auxiliary particle “saku” included in the input sentence.

[Supplement] The present invention has been described in detail with reference to the specific embodiments. However, it is obvious that those skilled in the art can modify or substitute the embodiments without departing from the scope of the present invention. That is, the present invention has been disclosed in the form of exemplification, and the contents of this specification should not be construed in a limited manner. In order to determine the gist of the present invention, the section of the claims described at the beginning should be taken into consideration.

[0062]

As described above in detail, according to the present invention,
It is possible to provide an excellent natural language processing system, natural language processing method, and computer program that can accurately determine the semantic information of a part of speech having a plurality of semantic functions.

Further, according to the present invention, an excellent natural language processing system and natural language processing method capable of accurately outputting the semantic information of a part of speech having a plurality of meanings depending on the method of use and a computer A program can be provided.

Further, according to the present invention, an excellent natural language processing system and natural language processing capable of accurately outputting the semantic information of a sub particle which may have a plurality of semantic functions depending on the appearance position. Methods as well as computer programs can be provided.

[Brief description of drawings]

FIG. 1 is a flowchart showing a processing procedure of a semantic function analysis of an auxiliary particle according to an embodiment of the present invention.

FIG. 2 is a diagram showing a description content of a sub particle “sword” in a dictionary.

FIG. 3 is a diagram showing an output of a semantic analysis result for an input sentence “He ate only three apples”.

FIG. 4 is a flowchart showing a processing procedure of semantic function analysis of an auxiliary particle according to another embodiment of the present invention.

FIG. 5 is a diagram showing a description content of a sub particle “up to” in a dictionary.

FIG. 6 is a diagram showing the output of the semantic analysis result for the input sentence “I ate an apple till him”.

FIG. 7 is a diagram schematically showing a configuration of a natural language processing system 1 based on LFG.

[Fig.8] Syntactically the input sentence "My daughter speaks English."
Fs obtained as a result of processing by the semantic analysis unit 1
It is the figure which showed structure.

[Figure 9] Syntactically the input sentence "My daughter speaks English."
Cs obtained as a result of processing by the semantic analysis unit 1
It is the figure which showed structure.

[Explanation of symbols]

1. Natural language processing system 2 ... Morphological analysis unit 2A ... Morphological rules, 2B ... Morphological dictionary 3 ... Syntactic and Semantic Analysis Department 3A ... Grammar rule, 3B ... Bond valence dictionary

Claims (17)

[Claims] 1. A natural language system for analyzing semantic information of a part of speech having a plurality of meanings, wherein a meaning candidate setting means for setting a meaning candidate of a part of speech having a plurality of meanings included in an input sentence, and an input sentence. Syntactic information extracting means for extracting the syntactic information of the part of speech included in, and meaning output means for selecting one of the meaning candidates as the meaning of the part of speech according to the extracted syntactic information of the part of speech. A natural language system characterized by: 2. The natural language processing system according to claim 1, wherein the part of speech having a plurality of meanings is an auxiliary particle. 3. The natural language processing system according to claim 1, wherein the meaning candidate setting means obtains candidates for the meaning of a part of speech by making an inquiry to a dictionary describing the meaning function of the part of speech. 4. The natural language processing system according to claim 1, wherein the syntactic information extraction means extracts a destination of the part of speech. 5. The natural language processing system according to claim 1, wherein the syntax information extracting means extracts a connection destination of the part of speech. 6. The meaning output means includes "degree" and "limitation".
If the syntactic derivation is a quantifier, the meaning of the auxiliary particle "just", which has two meanings, is determined to be "degree",
The natural language processing system according to claim 1, wherein "limited" is determined in other cases. 7. The meaning output means is "emphasized" and "degree".
When the sub particle "up to" that has two meanings is not connected to the case particle or the omitted case particle, and when it is connected to the omitted case particle "ni" or "he", The natural language processing system according to claim 1, wherein the meaning is determined to be "degree", and in other cases, it is determined to be "emphasis". 8. A method of selecting a candidate for meaning according to syntactic information of a part of speech is described in advance as a grammar rule, and the meaning output means selects the meaning of the part of speech according to the description of the grammar rule. The natural language processing system according to claim 1. 9. A natural language method for analyzing semantic information of a part of speech having a plurality of meanings, comprising a meaning candidate setting step of setting a candidate of a meaning in a part of speech having a plurality of meanings included in an input sentence, and an input sentence. A syntactic information extraction step of extracting the syntactic information of the part of speech included in the part of speech, and a meaning output step of selecting one of the meaning candidates according to the extracted syntactic information of the part of speech as the meaning of the part of speech. A natural language method characterized by: 10. A part-of-speech having a plurality of meanings is an auxiliary particle,
The natural language processing method according to claim 9, wherein. 11. The candidate for the meaning of part of speech is obtained in the step of setting the meaning candidate by obtaining an inquiry from a dictionary describing the meaning function of the part of speech.
Natural language processing method described in. 12. The natural language processing method according to claim 9, wherein in the syntactic information extracting step, a destination of the part of speech is extracted. 13. The natural language processing method according to claim 9, wherein in the syntax information extracting step, a connection destination of the part of speech is extracted. 14. In the meaning output step, a subposition particle “saku” having two meanings of “degree” and “limitation”,
10. The natural language processing method according to claim 9, wherein if the syntactically related destination is a quantifier, its meaning is determined to be “degree”, and otherwise, it is determined to be “limited”. 15. In the meaning output step, a sub particle “up to” having two meanings of “emphasis” and “degree” is omitted when it is not connected to a case particle or an omitted case particle. 10. When the case particle is connected to "ni" or "he", its meaning is determined as "degree", and in other cases, it is determined as "emphasis".
Natural language processing method described in. 16. A method of selecting a meaning candidate according to syntax information of a part of speech is described in advance as a grammar rule, and in the meaning output step, the meaning of the part of speech is selected according to the description of the grammar rule. The natural language processing method according to claim 9. 17. A computer program written in a computer-readable format so as to execute natural language processing for analyzing semantic information of a part of speech having a plurality of meanings on a computer system, the plurality of computer programs included in an input sentence. A meaning candidate setting step of setting a meaning candidate in a part of speech having the meaning of, a syntactic information extraction step of extracting the syntactic information of the part of speech included in the input sentence, and a meaning determination step according to the extracted syntactic information of the part of speech. And a meaning output step of selecting one of the candidates as the meaning of the part of speech.