JP2004171049A - Nature analysis system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system for much more objectively analyzing the nature of entity by removing the arbitrariness of the entity. <P>SOLUTION: In this nature analysis system, language issued from the entity is recognized concerning contents by a language recognition means 13. Also, a part of speech is extracted by a part of speech extraction means 14 from the recognized language obtained by the language recognition means 13 according to an "extraction algorithm". Furthermore, the nature of the entity is analyzed by a nature analysis means 15 according to an "analysis algorithm" based on the pattern of the extracted part of speech obtained by the part of speech extraction means 14. Then, information concerning the analyzed nature obtained by the nature analysis means 15 is provided to the entity or a third person by an information providing means 16. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a system for analyzing the nature of an entity based on a language emitted from the entity regarding the content.
[0002]
[Prior art]
The nature of an entity such as an organization such as a company or an organization or a person to which it belongs may be analyzed from various viewpoints by a third party from the viewpoint of some kind of rating or the right person in the right place. Further, for example, an objective analysis of a property of an entity is performed based on the concept of the entity for a certain material.
[0003]
[Problems to be solved by the invention]
However, in some cases, it is difficult to objectively analyze the nature of the entity because it wants the entity to be analyzed as having favorable characteristics, and presents an arbitrary idea that does not fit its own. It becomes.
[0004]
Therefore, an object of the present invention is to provide a system capable of analyzing the properties of an entity more objectively while eliminating the arbitraryness of the entity.
[0005]
[Means for Solving the Problems]
A property analysis system according to the present invention for solving the above-mentioned problem includes an extraction algorithm for extracting a part of speech from a language and an analysis algorithm for analyzing the property of the entity based on a part of speech pattern included in a language emanating from the entity with respect to the content. , Language recognition means for recognizing a language issued from an entity with respect to content, and part of speech extraction means for extracting a part of speech from a language recognized by the language recognition means in accordance with an extraction algorithm stored by the storage means Based on the part-of-speech pattern extracted by the part-of-speech extraction means, a property analysis means for analyzing the property of the entity according to the analysis algorithm stored by the storage means, and information on the property of the entity analyzed by the property analysis means, The entity or Characterized in that an information providing means for providing to a third party.
[0006]
According to the present invention, the nature of the entity is analyzed based on the part of speech pattern extracted from the language, not the language itself emitted from the entity. According to the knowledge obtained by the inventor of the present application, the "part-of-speech pattern" can reflect the nature of the entity more strongly than the language itself.
[0007]
Therefore, according to the present invention, even if the language of the content emitted from the entity is arbitrary, the nature of the entity can be objectively analyzed by eliminating the arbitraryness.
[0008]
Further, the property analysis system according to the present invention is characterized in that the storage means stores the content, and includes a content providing means for providing the entity with the content stored by the storage means.
[0009]
According to the present invention, a series of processes of providing content, recognizing a language emitted from an entity with respect to the content, extracting part of speech from this language, and analyzing the properties of the entity based on the extracted part of speech pattern are comprehensively performed. Can be managed.
[0010]
Further, the property analysis system of the present invention is characterized in that the property analysis means parameterizes and analyzes the property of the entity based on the part of speech pattern extracted by the part of speech extraction means.
[0011]
According to the present invention, the nature of an entity can be objectively analyzed through “parameters” that reflect patterns of parts of speech extracted from languages uttered from the entity with respect to content.
[0012]
Further, the property analysis system of the present invention includes state detection means for detecting the state of the entity in a time-series manner in correspondence with the language issued by the entity in accordance with the content, and the property analysis means is extracted by the part-of-speech extraction means. Based on the part-of-speech pattern detected and the state of the entity detected by the state detection unit in time series correspondence with the part-of-speech pattern, the property of the entity is analyzed according to the analysis algorithm stored in the storage unit. It is characterized by the following.
[0013]
According to the present invention, the nature of the entity is analyzed based on the part-of-speech pattern extracted from the language emitted from the entity with respect to the content, and the state of the entity that is time-correlated with the part-of-speech pattern. According to the knowledge obtained by the inventor of the present application, the "part of speech pattern" and the "state" of an entity that is time-sequentially associated with this part of speech pattern reflect the nature of the entity hidden behind the language more strongly. obtain.
[0014]
Therefore, according to the present invention, even if the language of the content emitted from the entity is arbitrary, it is possible to eliminate this arbitraryness and analyze the nature of the entity more objectively.
[0015]
Further, in the property analysis system according to the present invention, the property analysis unit converts the property of the entity into a primary parameter based on the part-of-speech pattern extracted by the part-of-speech extraction unit, and is associated with the part-of-speech pattern in time series by the state detection unit. The sum of the weighted primary parameters according to the state of the detected entity is calculated as a secondary parameter representing the property of the entity.
[0016]
According to the present invention, in addition to a part-of-speech pattern extracted from a language emanating from an entity with respect to content, a “secondary parameter” that reflects the state of an entity that is associated with the part-of-speech pattern in chronological order is reflected. The nature of the entity can be analyzed objectively.
[0017]
Further, in the property analysis system of the present invention, the storage means stores information on the existing, historical or fictional person or group in association with the pattern of the property of the entity, and the information providing means analyzes the entity analyzed by the property analysis means. According to the characteristic pattern, information about a person or an organization stored in the storage means is provided to the entity or a third party.
[0018]
In accordance with the present invention, the nature of an entity can be projected onto historical, existing or fictional persons or entities, revealed, and analyzed.
[0019]
BEST MODE FOR CARRYING OUT THE INVENTION
An embodiment of the property analysis system of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating the configuration of the property analysis system according to the present embodiment, and FIGS. 2 and 3 are diagrams illustrating the function of the property analysis system according to the present embodiment.
[0020]
The property analysis system shown in FIG. 1 includes a server 10 capable of network communication with a terminal 20 of an "entity" to be analyzed.
[0021]
Further, the present system includes a storage unit 11, a content providing unit 12, a language recognizing unit 13, a part of speech extracting unit 14, a property analyzing unit 15, and an information providing unit 16.
[0022]
The storage unit 11 stores contents, an extraction algorithm, an analysis algorithm, and the like by an electromagnetic method. The storage unit 11 stores information on existing, historical, or imaginary persons or organizations in association with the pattern of the nature of the entity by an electromagnetic method.
[0023]
The content providing unit 12 provides the “content” stored in the storage unit 11 to the entity via the network and the terminal 20. That is, the “content” is downloaded from the server 10 to the terminal 20 by the function of the content providing unit 12.
[0024]
The language recognizing unit 13 recognizes the content by acquiring the language input by the entity to the terminal 20, that is, the language issued from the entity via the network. That is, the “language” of the entity is uploaded from the terminal 20 to the server 10 by the function of the language recognition unit 13.
[0025]
The part-of-speech extraction unit 14 extracts “part-of-speech” from the “language” recognized by the language recognition unit 13 according to the “extraction algorithm” stored in the storage unit 11.
[0026]
The property analysis unit 15 analyzes the property of each entity based on the “part of speech pattern” extracted by the part of speech extraction unit 14 according to the “analysis algorithm” stored in the storage unit 11. The property analysis means 15 parameterizes the property of the entity.
[0027]
The information providing means 16 provides information on the properties of the entity analyzed by the property analyzing means 15 to the entity or the like. That is, by the function of the information providing means 16, information on the nature of the entity is downloaded from the server 10 to the terminal 20.
[0028]
The entity terminal 20 has various functions such as an information processing function by a CPU and the like, a network communication function, a display function of contents and the like by a monitor, and an input operation function of a language and the like by a keyboard and a mouse.
[0029]
The function of the property analysis system having the above configuration will be described with reference to FIGS.
[0030]
First, "content" is downloaded from the server 10 to the terminal 20 of each entity (arrow (1) in FIG. 2). Examples of the "content" include anecdotal words such as "crying and slashing", a story such as "three piglets", a still image such as a battlefield scene, and a movie such as a movie with a story. This causes the terminal 20 to display the content and a message requesting that the entity present an opinion on the content. The content is preferably a well-known or well-known content so that the entity can be expected to have some prior knowledge and various opinions can be presented. Also, it is preferable that the message is a vague question so that a multifaceted opinion can be drawn from the entity.
[0031]
Next, the entity operates the terminal 20 to respond to the message with an opinion corresponding to the content, such as "〜 is XX. However,-is ●. Therefore,-is ‥." Enter in the form of a language. Thereby, the opinion (language) issued from the entity regarding the content is uploaded from each terminal 20 to the server 10 (arrow (2) in FIG. 2). Note that, after a plurality of options are presented, an opinion of the entity may be input by selecting one of the options by a click operation or the like by the entity.
[0032]
Subsequently, the part-of-speech extraction unit 14 extracts, from the language uploaded from the terminal 20 to the server 10, “part-of-speech” that is a class of words classified according to grammatical properties according to the “extraction algorithm” stored in the storage unit 11. Extract (FIG. 2s1). The "part of speech" extracted by the part of speech extraction means 14 includes a noun, a pronoun, an adjective, an adjective verb, a verb, an adverb, a conjunction, a verb, an auxiliary particle, an auxiliary verb, an adverb, and the like. Note that, depending on the difference in the language system, parts of speech different from the above may be extracted, or these parts of speech may be classified in more detail or more comprehensively and then extracted by the part of speech extracting means 14.
[0033]
In addition, the property analysis unit 15 analyzes the property of the entity based on the part-of-speech pattern extracted by the part-of-speech extraction unit 14 according to the “analysis algorithm” stored in the storage unit 11 (s2 in FIG. 2). Specific examples of the part-of-speech pattern include the appearance frequency and type of predetermined parts of speech, such as nouns, conjunctions, and conjunction particles, and whether they are logically matched or not. Also, as specific examples of the nature of the entity, “philosophy view” is XX, from “honesty view” is ●●, “business genius” is ◎◎, “patience” is ~, And properties from various viewpoints such as 挙 げ.
[0034]
In addition, the property analysis means 15 determines that the philosophy index is “α”, the sincerity index is “β”, the genius index is “γ”, the patience index is “δ”, and the mask index (the degree to which emotions and thoughts are not displayed on the surface). ) Is parameterized according to the analysis algorithm stored in the storage unit 11 such as “ε” and ‥ (see s2).
[0035]
Further, the information providing means 16 determines that the philosophy index is “α”, the sincerity index is “β”, the merchandising index is “γ”, the patience index is “δ”, the masking index is “ε”, and ‥, etc. According to the pattern, the storage unit 11 reads information about a historical person or the like stored in association with the pattern of the property. Specific examples of information about a person or the like include information such as "Nobunaga Oda type has diversified management ability, ‥" or "Genius who is willing to spare Edison type effort, ‥". It is preferable that a person or the like included in this information is a well-known or well-known person or the like that can make the entity feel familiar. Further, the content describing the person or the like included in this information does not necessarily have to be accurate, and may be content that is generally believed.
[0036]
Then, information relating to the property of the entity and information relating to a person or the like according to the pattern of the property are downloaded from the server 10 to the terminal 20 of the entity (arrow {3} in FIG. 2).
[0037]
Thereby, windows 31 to 33 shown in FIG. 3 are displayed on the monitor of terminal 20. In the window 31, “philosophy view” is XX, and information about the property of the entity such as Δ is displayed. Further, in the window 32, parameters representing each property of the entity, such as a "merchant index" and a "mask index", are displayed as a waveform graph in which the amplitude is emphasized by taking the index. Note that these parameters may be represented by a bar graph, a regular polygon graph, or the like. Further, the window 33 displays information about a historical person or the like, such as "Nobunaga Oda type has diversified management ability," corresponding to the pattern of the nature of the entity. The window 33 also displays a portrait or cartoonized character such as the person on the terminal 20.
[0038]
Note that these pieces of information may be downloaded from the server 10 to the terminal 20 of another entity (third party) other than the entity itself, and provided to the other entity.
[0039]
According to this system, the nature of an entity is analyzed based on the part-of-speech pattern extracted from the opinion, not the opinion (language) of the entity itself according to the content. It is thought that the “part of speech pattern” can reflect the nature of the entity more strongly than the opinion itself. For example, suppose that the entity has expressed an opinion that "it must be noted that it does not become a four-sided song" with respect to the predecessor word "seven-sided song". From this opinion itself, it seems that the entity's view of philosophy and life is superior. However, even if the opinion itself is admirable, the part-of-speech pattern contained in this opinion indicates that it is inferior in properties such as genius and perseverance that seem to be completely unrelated to the factual word. Is possible.
[0040]
Therefore, according to the present system, even if the language of the content emitted from the entity is arbitrary, the nature of the entity can be objectively analyzed by removing the arbitraryness.
[0041]
In addition, the entity can grasp the analysis result of the property of itself or another entity via windows 31 to 33 displayed on the terminal 20 (see FIG. 3). That is, the entity is "OO" in the "philosophy view" of the target entity through the window 31 displayed on the terminal 20, is "●" from "Honesty view", "commercial" is "◎◎", and ""Patience" is ~, and it is possible to grasp properties from various viewpoints such as ‥.
[0042]
Further, the entity can grasp a parameter representing the property via the window 32 displayed on the terminal 20. That is, the entity can specifically grasp the somewhat vague analysis result in the abstract language through the size of the parameter.
[0043]
Further, the entity can project the target entity to a historical person or the like via the window 33 displayed on the terminal 20, and can grasp the outline of the property of the target entity in a familiar and easy-to-understand manner. As described above, the device for visually grasping the information on the property of the entity is enhanced, so that the use of the present system becomes more attractive and the system can be widely spread.
[0044]
In the present embodiment, the property analysis system is configured by the server 10, but as another embodiment, the property analysis system may be configured by the terminal 20 of the entity. According to the other embodiment, the content and the algorithm are stored in the HDD (storage device / not shown) of the terminal 20, and the information and the like on the property of the content and the entity are stored in the terminal 20 by the function of the CPU or the like according to the operation of the entity. Is displayed on a monitor (not shown) (see FIG. 3).
[0045]
Further, in the present embodiment, the “content” is provided to the entity via the network by the content providing means 12 (see arrow {circle around (1)} in FIG. 2), but as another embodiment, paper, CD-ROM, video tape, etc. "Content" may be provided to entities by distribution of various recording media. In the other embodiment, the content providing unit 12 may be omitted.
[0046]
Further, in the present embodiment, the "opinion (language)" for the content is provided from the entity via the network (see the arrow {circle around (2)} in FIG. 2), but in another embodiment, the "opinion" is transmitted via a recording medium such as paper. May be provided by the entity. In this other embodiment, the language may be recognized by an optical method, such as the language recognizing unit 13 irradiating light onto a recording medium such as paper and then recognizing the language of the entity based on the reflected light.
[0047]
Further, in the present embodiment, the information providing means 16 provides the information about the nature of the entity, the information about the historical person, etc. to the entity via the network (see the arrow {circle around (3)} in FIG. 2). In another embodiment, such information may be provided to the entity via a recording medium such as paper or a CD-ROM. In the other embodiment, the information providing unit 16 may cause the information recording device to record or record the information on the recording medium, thereby issuing or creating a recording medium on which the information is recorded.
[0048]
The property analysis system includes “state detection means (not shown)” that detects the state of the entity in a time-series manner in correspondence with the language issued by the entity in accordance with the content. Based on the part-of-speech pattern extracted by the method and the state of the entity detected by the state detection unit in time series correspondence with the part-of-speech pattern according to the analysis algorithm stored in the storage unit 11. May be analyzed.
[0049]
According to such a configuration, the nature of the entity is analyzed based on the part-of-speech pattern extracted from the language emanating from the entity with respect to the content, and the state of the entity associated with the part-of-speech pattern in chronological order (FIG. 2s2). Specific examples of the “state” of the entity include the frequency of the arm or leg of the entity, body temperature, heart rate, amount of sweat, wavelength and amplitude of brain waves, and the like. It is considered that the nature of the entity hidden behind the language can be more strongly reflected in the part-of-speech pattern and the state of the entity that is associated with the part-of-speech pattern in chronological order. Therefore, even if the language of the content emitted from the entity is arbitrary, the nature of the entity can be further objectively analyzed by eliminating the arbitraryness.
[0050]
Further, the property analysis unit 15 converts the property of the entity into a primary parameter based on the part-of-speech pattern extracted by the part-of-speech extraction unit 14, and is associated with the part-of-speech pattern in time series by a state detection unit (not shown). The sum of the weighted primary parameters according to the detected state of the entity may be calculated as the secondary parameter representing the property of the entity.
[0051]
In the present embodiment, specific examples of the nature of the entity include “philosophy view”, “sincerity view”, “commercial talent”, “patience”, “common sense”, “leadership”, and the like. ) Were analyzed for these properties, but in another embodiment, as specific examples of the properties of the entity, "logical thinking power", "concepts for small and medium-sized concepts", "recognition of timekeeping change creation", "question answer" Intentional ability, "Sustained question answering,""Systemcomprehension,""Retention of memory,""Languagesense," etc. These properties may be analyzed for the entity).
[0052]
Here, "logical thinking ability" refers to the ability to have consistent thoughts through muscles on the premise of conditions, "concept / small comprehension ability" refers to the ability to grasp conceptual / abstract contents, "Time-based change creation recognition" refers to the ability to create scene changes, progress of chemical reactions, predictions of changes in the human mind, rotation of figures, etc., and "problem-answering ability" refers to the ability to see questions and obtain clues to answers. And "persistence in answering questions" refers to the persistence (persistence in a good way) to derive answers to questions, and "systematic understanding" refers to the ability to systematically understand or understand a certain learning field or matter. "Memory retention" indicates the ability to use knowledge learned in the past to derive the answer to the question, and "language sense" indicates the appropriateness of the grammar and logic development used.
[0053]
According to the other embodiment, the entity (examinee or its instructor) determines that the “logical thinking power” of the target entity (examinee) is △△ via the windows 31 to 33 displayed on the terminal 20; “Concept / Small / Medium Understanding Ability” is ▲▲, and “Timekeeping Change Creation Recognition” can grasp the nature of the entity from various viewpoints such as ~.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating the configuration of a property analysis system according to an embodiment; FIG. 2 is a diagram illustrating the function of a property analysis system according to the embodiment; FIG. 3 is a diagram illustrating the function of a property analysis system according to the embodiment;
10 server, 11 storage means, 12 content provision means, 13 language recognition means, 14 part of speech extraction means, 15 property analysis means, 16 information provision means, 20 terminal of entity

Claims (6)

Storage means for storing an extraction algorithm for extracting a part of speech from a language, and an analysis algorithm for analyzing a property of the entity based on a pattern of a part of speech included in a language issued from the entity with respect to the content;
A language recognizer for recognizing the language of the content emitted by the entity;
A part-of-speech extraction unit that extracts a part-of-speech from a language recognized by the language recognition unit according to an extraction algorithm stored by the storage unit;
Based on the part-of-speech pattern extracted by the part-of-speech extraction means, a property analysis means for analyzing the property of the entity according to an analysis algorithm stored by the storage means;
A property analysis system, comprising: information providing means for providing information on the property of an entity analyzed by the property analysis means to the entity or a third party. 2. The property analysis system according to claim 1, wherein the storage means stores the content, and further comprises content providing means for providing the content stored by the storage means to the entity. 3. The property analysis system according to claim 1, wherein the property analysis means parameterizes and analyzes the property of the entity based on the part of speech pattern extracted by the part of speech extraction means. State detecting means for detecting the state of the entity in a time-series manner in correspondence with the language issued by the entity according to the content,
It is stored by the storage unit based on the part-of-speech pattern extracted by the part-of-speech extraction unit by the property analysis unit and the state of the entity detected by the state detection unit in chronological correspondence with the part-of-speech pattern. 4. The property analysis system according to claim 1, wherein the property of the entity is analyzed according to an analysis algorithm. The property analyzing unit converts the property of the entity into a primary parameter based on the part-of-speech pattern extracted by the part-of-speech extracting unit, and according to the state of the entity detected by the state detecting unit in time series with the detected part-of-speech pattern. 5. The property analysis system according to claim 4, wherein the sum of the weighted primary parameters is calculated as a secondary parameter representing the property of the entity. Storage means for storing information about existing, historical or fictional persons or entities in association with the pattern of the nature of the entity;
Claims characterized in that the information providing means provides the information on the person or the organization stored in the storage means to the entity or a third party according to the pattern of the property of the entity analyzed by the property analyzing means. Item 6. The property analysis system according to item 1, 2, 3, 4, or 5.

Images