JP4922252B2 - Effect expression presentation apparatus and method - Google Patents

Description

  The present invention relates to an effect expression presentation device and a method thereof. In particular, the present invention relates to a mechanism for presenting effective expressions for words input by a user.

Products and services that are sold and offered, and things such as events are given names to identify them. These names (product naming) are the modern times when information transmission means such as television and the Internet have developed. Has a great advertising function. As a result, commodity naming, has led to a significant impact on the success of the products and services of sales, or event. Here, in the current situation, product naming with high influence often has similar readings (sounds) to the product. For this reason, there is a need for a mechanism that allows a person to easily come up with an effective expression that is easily familiar to society, that is, similar in reading to the product.

Incidentally, the information from the user conventionally is input, the information presentation device for presenting known information corresponding to the input information. In such an information presentation device, there is a device that makes it difficult for the user to feel bored by using so-called “bokeh” or “tsukkomi” seen in comics or the like as information to be presented (see Patent Document 1). According to such an information presentation apparatus, information rich in variations that has not been available can be presented with respect to input information from the user.
JP 2006-268428 A

  However, the information presentation apparatus as described above can only present information according to a scenario even though it is rich in information, and cannot present an effective expression for a certain thing.

  Therefore, the present invention provides a mechanism that can provide an effective expression for user input.

  The present invention provides the following solutions.

(1) An effect expression presentation device that presents an effective expression for a keyword input by a user, a reading extraction unit that extracts a character string indicating the reading of the keyword, and a synonym for the keyword A synonym extraction unit that extracts a character string indicating the reading of the keyword, an edit distance calculation unit that calculates an edit distance indicating a degree of approximation between the character string of the keyword and the extracted character string of the synonym, and the calculated edit distance and based on whether the part of speech of the part of speech and the synonyms of the keywords match, and effectively represent the score calculation unit for calculating a score for the keyword of the synonyms, presented in association with the score calculated the synonyms An effect expression presentation device comprising: an effect expression presentation unit.

According to the configuration of (1), the edit distance calculation unit calculates an edit distance indicating the degree of approximation between the keyword character string and the extracted synonym character string. Then, the effect expression score calculation unit calculates a score for the synonym keyword based on the calculated edit distance and whether or not the keyword part of speech and the part of speech of the synonym match, and the effect expression presentation unit calculates the synonym Present in association with the score.

  Thus, the score is calculated based on the degree of approximation (edit distance) between character strings and whether or not the parts of speech match. Here, when the parts of speech are different, it is highly valuable as an effect expression. Therefore, it is possible to determine the degree of effect for the input keyword from the score for the presented synonym. From the user's point of view, by confirming the presented synonyms and their scores, an effective expression for the keyword can be easily found, so that an effective expression for the user's input can be provided.

  (2) The effect expression presentation device according to (1), wherein the synonym extraction unit extracts a semantic synonym having a similar meaning to the keyword and a nominal synonym having a similar designation to the keyword. A feature effect presentation device.

According to the configuration of (2), not only a semantic synonym but also a synonym synonym is used as a synonym, so that an unexpected synonym may be extracted for a keyword. As a result, an effective expression for the user input can be provided.

  (3) The effect expression presenting device according to (1) or (2), wherein the effect expression presenting unit presents only synonyms whose scores are within a predetermined range. .

  According to the configuration of (3), since there are synonyms that are not presented depending on the score, only synonyms that are highly effective are presented to the user. As a result, it is not necessary for the user to check even useless synonyms, and an effective expression can be easily found.

  (4) The effect expression presentation device according to (3), further comprising a threshold setting unit configured to set the predetermined range.

  According to the configuration of (4), since the user can set the predetermined range, it is possible to present effective expressions according to various cases.

(5) A method of presenting an effective expression for a keyword input by a user, the step of extracting a character string indicating the reading of the keyword, and a character indicating the reading of the synonym for the synonym for the keyword Extracting with a sequence; calculating an edit distance indicating a degree of approximation between the character string of the keyword and the extracted character string of the synonym; the calculated edit distance, the part of speech of the keyword, and the part of speech of the synonym Doo is based on whether the match, the method comprising the steps of calculating a score for the keyword of the synonyms, and presenting in association with the score calculated the synonyms, a.

  According to the configuration of (5), it is possible to provide an effective expression for the user input as in the configuration of (1).

  ADVANTAGE OF THE INVENTION According to this invention, the effective expression with respect to a user's input can be provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing an example of the configuration of an effect expression presentation device 10 of the present invention.

  The effect expression presentation device 10 includes a control unit 100 and a hard disk 200. The control unit 100 includes a user input receiving unit 11, a reading extraction unit 12, a semantic synonym extraction unit 13, a nominal synonym extraction unit 14, an edit distance calculation unit 15, an effect expression score calculation unit 16, and an effect expression presentation. Part 17. The hard disk 200 stores a semantic synonym dictionary 21 and a nominal synonym dictionary 23.

  The user input receiving unit 11 receives input of keywords from the user. The reading extraction unit 12 extracts the reading of the input keyword. Here, in the present invention, a character string in Roman alphabet is used as a reading. The semantic synonym extraction unit 13 and the nominal synonym extraction unit 14 extract the synonyms of the input keyword from the semantic synonym dictionary 21 and the nominal synonym dictionary 23, respectively.

  Here, as shown in FIGS. 3 and 4, the semantic synonym dictionary 21 and the nominal synonym dictionary 23 define a plurality of words for each predetermined type. The semantic synonym extraction unit 13 and the nominal synonym extraction unit 14 extract the type of word to which the input keyword belongs as a synonym. An input keyword may belong to only one type or may belong to a plurality of types.

The edit distance calculation unit 15 calculates the edit distance between the input keyword and the extracted synonym. Here, the edit distance is the degree of approximation between the input keyword string and the synonym string. Character replacement / insertion to make the input keyword string a synonym string -The minimum number of deletions. For this reason, when the character string of the input keyword and the character string of the synonym completely match, the editing distance becomes zero. Note that the method of calculating the edit distance will be described with reference to FIGS.

The effect expression score calculation unit 16 calculates a score indicating the degree of effect of the synonym on the keyword based on the edit distance calculated by the edit distance calculation unit 15 and the match / mismatch between the part of speech of the keyword and the part of speech of the synonym. The effect expression presenting unit 17 presents the calculated score to the user in association with the synonym. Incidentally, the calculated score may be to present to the user only a predetermined range der Ru s Gigo.

  In the present invention, the input keyword may be a word or a sentence. Here, when the input keyword is a sentence, a score can be calculated after performing a morphological analysis on the sentence. Therefore, in the present invention, the morphological analysis unit 18 may be provided as an option when the input keyword is a sentence.

  FIG. 2 is a diagram illustrating an example of a hardware configuration of the present invention. The effect expression presentation device 10 includes a central processing unit (CPU) 1010 (a plurality of CPUs such as CPU_A 1012 may be added in a multiprocessor configuration) constituting a control unit 100, a bus line 1005, a communication I / F 1040, a main memory 1050. , BIOS (Basic Input Output System) 1060, hard disk 200, keyboard and mouse 1100 and other input means and display device 1022.

  The BIOS 1060 stores a boot program executed by the CPU 1010 when the effect expression presentation device 10 is started up, a program depending on hardware, and the like.

  The hard disk 200 stores various programs for the effect expression presentation device 10 to function and programs for executing the functions of the present invention, and can configure various databases as necessary.

The display device 1022 displays a screen for accepting data input to the user or displays a calculated effect expression phrase, and a display device such as a cathode ray tube display device (CRT) or a liquid crystal display device (LCD). including.

  Here, the input means accepts input by the user, and is configured by a keyboard, a mouse 1100, and the like.

The communication I / F 1040 is a network adapter for enabling the effect expression presentation device 10 to be connected to another device via a dedicated network or a public network.

  Next, the semantic synonym dictionary 21 and the nominal synonym dictionary 23 stored in the hard disk 200 will be described with reference to FIGS. 3 and 4.

The semantic synonym dictionary 21 defines a plurality of words having meanings belonging to a predetermined concept, that is, a plurality of words for each predetermined type. The word, the reading pseudonym, to read consists of a string of goods Shi及 beauty romanization data are defined.

The nominal synonym dictionary 22 defines a plurality of words having a certain similar designation, that is, a plurality of words for each predetermined type. The word, the reading pseudonym, to read consists of a string of goods Shi及 beauty romanization data are defined.

  Next, referring to FIG. 5, a flowchart showing the processing contents of the present invention will be described.

  First, in step S101, the control unit 100 receives an input of a keyword from a user. Specifically, keyword input from the keyboard 1100 is received and stored in the main memory 1050.

  Next, in step S102, the control unit 100 performs morphological analysis on the input keyword and extracts a noun. Note that this processing can be omitted when the input keyword is a word.

  Next, in step S103, the control unit 100 extracts keyword reading data, that is, a character string in Roman alphabet.

  Next, in step S104, the control unit 100 identifies a semantic type to which the input keyword belongs, and extracts words belonging to this type. Further, in step S105, the name classification to which the inputted keyword belongs is specified, and the words belonging to this classification are extracted.

Next, in step S106, the control unit 100 calculates an edit distance between the input keyword and the extracted word. Thus, in the present invention, only words that are similar in meaning or name to the input keyword are targeted for the effect expression phrase. Therefore, it is possible to reduce the control load as compared with the case where all words are targeted, and it is possible to extract words that are more effective.

  Here, referring to FIG. 6 to FIG. 8, an editing distance calculation method using an example in which the input keyword is “futon” and the extracted word is “hutbubu”. explain.

For the calculation of the edit distance, a matrix composed of the character string of the input keyword and the character string of the extracted word is used. In this matrix , a value obtained by adding one by one from the first character to the last character of the character string is assigned in advance (initial state). FIG. 6 is a diagram showing an initial state in calculation of the edit distance. Thereafter, the edit distance is calculated by filling the blank portion in the initial state according to a predetermined rule.

Specifically, as shown in FIG. 7, the upper left blank part includes (1) the upper left value of this blank + replacement cost, (2) the left value of this blank + 1, and (3) the upper part of this blank. The smallest value among the values of +1 is filled. Note that substitution cost is becomes "0" when the the character of a word and the extracted character keyword corresponding to this column is identical, if they do not match is "1".

  The blank part is filled under such a rule, and finally, the smallest value in the lowest column is the edit distance between the input keyword and the extracted word. Here, the editing distance is 2 between the keyword “futon” and the extracted word “blow away” (see FIG. 8).

  Referring to FIG. 5 again, in step S107, control unit 100 calculates an effect score from the calculated edit distance, part of speech, and the like. This process is performed by first calculating an effect score based on the calculated edit distance, and then correcting the effect score by matching / disagreement of part of speech.

The calculation of the effect score based on the edit distance is, for example, “1” when the edit distance is equal to or greater than ½ of the length of the character string in the Roman alphabet of the keyword. That is, when the keyword is “futon”, the length of the character string “huton” written in Roman letters is “5”. In this case, when the edit distance is “3” , the edit distance is ½ or more of the length of the character string in the Roman alphabet of the keyword.

  Similarly, when the editing distance is ¼ or more and less than ½ of the length of the character string in the Roman alphabet of the keyword, “3” is set. For this reason, the effect score based on the above-mentioned “blow away” editing distance is “3”. When the edit distance is greater than 0 and less than ¼, “5” is set, and when the edit distance is 0, “7” is set.

  When the keyword and the extracted word are the same, the edit distance is zero. However, the same word as the keyword cannot be an effective expression for the keyword. Therefore, the effect score may be set to “0” for the same word as the keyword. Furthermore, it is preferable that the same word as the keyword is not extracted as a synonym. Note that different notation may be made even for the same word. For example, “Futon” is written as “Futan”, “Futon”, and “Futon”. Even in such a case, it is preferable that each word is not extracted as a synonym. Therefore, in the semantic synonym dictionary 21 and the nominal synonym dictionary 23, in addition to “reading kana”, “part of speech”, “reading data”, a “semantic identifier” is provided, and words having the same “semantic identifier” are not extracted. Can be.

  By doing so, for example, when the keyword is “futon”, it is possible not to extract “Kashidan” etc., but when the keyword is “Shiga”, it is effective such as “tooth” etc. It is possible to extract words that can be expressed easily.

  When the effect score based on the edit distance is calculated in this way, the effect score is corrected next by the coincidence / mismatch of the parts of speech. For example, “1” can be added to the effect score when the parts of speech are different. If the parts of speech are different, the value to be added can be changed depending on the size of the edit distance, and the value to be added can be changed depending on the part of speech, instead of adding “1” uniformly. For example, “2” is added when the edit distance is ¼ or more and less than ½ of the length of the character string written in Romaji, but not added when the edit distance is “0”. Good. Further, “2” may be added when a word whose keyword is extracted as a noun is a verb, and “1” may be added when a word whose keyword is extracted as a noun is an adjective.

  The reason for correcting the score in the part of speech is that, in general, when the part of speech is different, the value becomes high as an effect expression. As will be described later, the calculation of the score based on the edit distance and the correction of the score based on the part-of-speech information may differ depending on the user's application.

  Referring to FIG. 5 again, in step S108, control unit 100 performs threshold determination for selecting only synonyms whose calculated scores are within a predetermined range. Next, in step S109, the control unit 100 presents the word selected in step S108 on the display device 1022, and ends the process.

  As described above, in the present invention, not all the extracted words but only words having a score within a predetermined range are presented to the user, so that the user does not need to check even useless words and is more effective. Words can be easily found.

  This predetermined range may be determined in advance by experience, or may be arbitrarily set by the user using input means such as a keyboard and mouse 1100. In particular, when the user can arbitrarily set it, only effective words according to the user can be presented, and the user's needs can be further met.

  Next, display examples in the display device 1022 of the present invention will be described with reference to FIGS.

First, the keyword input acceptance screen will be described with reference to FIG. On the display screen of the display device 1022 , a keyword input acceptance column 2010 and a usage acceptance column 2050 are displayed. The keyword input reception column 2010 receives an input of a keyword from the user. The use acceptance column 2050 accepts the purpose for the user to use the effect expression presentation device 10 according to the present invention. Note that the display device 1022 also displays a start reception unit that triggers the presentation of the effect expression, but is not shown in FIG.

  In the usage acceptance column 2050, a plurality of predetermined usages are displayed so as to be selectable. These uses are provided so that the ranges of scores when presented are different. The user can present scores in different ranges by selecting the displayed application. Uses that can be selected may include not only different score ranges but also different score calculation methods. That is, depending on the application, the difference in part of speech may be important. In such an application, the correction of the score due to the difference in part of speech may be increased. Depending on the application, it may not be appropriate if the edit distance is too small or too large. In such a case, it may be preferable to change the score for the edit distance depending on the application.

Next, the effect expression presentation screen will be described with reference to FIG. On the display screen of the display device 1022, a keyword input acceptance column 2010 and an effect expression presentation column 2070 are displayed. Keywords input by the user are displayed in the keyword input reception column 2010 , and effective words selected by a series of processes are displayed in the effect expression presentation column 2070. Here, the display of the effect expression presentation column 2070 includes a use 2071, a word 2073, and a score 2075. The usage 2071 displays the usage selected by the user. Word 2073 and score 2075 display the selected word and its score.

  In this way, since effective words are displayed in association with the calculated score, the user can recognize the objective effect level for the keyword, and as a result, the effective word for the keyword can be easily obtained. Can be found.

Here, the use expression 2071 and the score 2075 are displayed in the effect expression presentation field 2070. Therefore, when the score calculation method is different depending on the usage that can be selected, the user grasps the score 2075 in the displayed word 2073 so that the usage (that is, the score calculation method) selected by the user is the same. It can be recognized whether it was appropriate or not. As a result, even a user have become familiar Lena in operation and attempting repeatedly, what kind of words will be able to figure out the selected easily according to the application, facilitating the effective words which the user desires To be able to find.

  Also, in order to select effective words, not only semantic synonyms but also nominal synonyms are used. Therefore, there is a possibility that a word that does not come to mind about the keyword, that is, a word that has a different meaning but has a similar name, can be selected, and a word that is more effective for the user can be selected. In this case, since only the synonyms of the input keyword are targeted, the control load can be reduced as compared to the case of targeting all words, and more effective words can be extracted. it can.

In addition, since only words according to usage are selected, unnecessary words that are meaningless to the user are not displayed, and the user can easily find effective words without making unnecessary checks. It becomes possible . In the case of this, since you are able to be the application user to set, it is possible to display the effective words corresponding to various cases.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

  For example, the semantic synonym dictionary 21 and the nominal synonym dictionary 23 may be sequentially updated by user input. That is, in the semantic synonym dictionary 21 and the nominal synonym dictionary 23, a word whose part of speech is a verb is also registered. The user can add a utilization form of the verb later.

It is a figure which shows the effect expression presentation apparatus of this invention. It is a figure which shows the hardware constitutions of the effect expression presentation apparatus of this invention. It is a figure which shows the semantic synonym dictionary of this invention. It is a figure which shows the nominal synonym dictionary of this invention. It is a flowchart which shows the processing content of the effect expression presentation apparatus of this invention. It is a figure which shows the calculation method of edit will distance. It is a figure which shows the calculation method of edit will distance. It is a figure which shows the calculation method of edit will distance. It is a figure which shows the example of a display of the input reception screen of the effect expression apparatus of this invention. It is a figure which shows the example of a display of the effect expression presentation screen of the effect expression apparatus of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Effect expression presentation apparatus 11 User input reception part 12 Reading extraction part 13 Semantic synonym extraction part 14 Nominal synonym extraction part 15 Edit distance calculation part 16 Effect expression score calculation part 17 Effect expression presentation part 100 Control part 200 Hard disk

Claims (5)

An effect expression presentation device for presenting an effective expression for a keyword input by a user,
A reading extractor for extracting a character string indicating the reading of the keyword;
A synonym extraction unit for extracting a synonym for the keyword together with a character string indicating a reading of the synonym;
An edit distance calculating unit for calculating an edit distance indicating a degree of approximation between the character string of the keyword and the extracted character string of the synonym;
An effect expression score calculating unit that calculates a score for the keyword of the synonym based on whether the calculated edit distance and the part of speech of the keyword and the part of speech of the synonym match;
An effect expression presentation unit that presents the synonym in association with the calculated score;
An effect expression presentation device comprising: The effect expression presentation device according to claim 1,
The synonym extraction unit extracts a semantic synonym having a similar meaning to the keyword and a nominal synonym having a similar name to the keyword. The effect expression presentation device according to claim 1 or 2,
The effect expression presentation unit presents only synonyms whose scores are within a predetermined range. The effect expression presentation device according to claim 3,
An effect expression presenting device, further comprising a threshold setting unit for setting the predetermined range. A computer comprising a control unit and a storage unit is a method for presenting an effective expression for a keyword input by a user,
The control unit executes according to a program stored in the storage unit,
Extracting a character string indicating the reading of the keyword;
Extracting a synonym for the keyword along with a character string indicating a reading of the synonym;
Calculating an edit distance indicating a degree of approximation between the keyword character string and the extracted synonym character string;
Calculating a score for the keyword of the synonym based on the calculated edit distance and whether the part of speech of the keyword and the part of speech of the synonym match;
Presenting the synonym in association with the calculated score;
Including methods.

Images