Term detecting device and program

Abstract

To reduce, when a term that needs a caution is detected from a text, a process load of such a detection, and to present, to a user, the detection result in an easy-to-understand form.SOLUTION: A morphological analysis unit 12 of a telop production assisting device 2 performs a morphological analysis on an input text, and a verb replacing unit 13 replaces an inflected form of a verb contained in the text having undergone the morphological analysis with an original form. Caution-for-use term detecting means 21 performs pattern matching between the text having undergone the verb replacement and caution-for-use term information 32, detects a caution-for-use term from the text having undergone the verb replacement, generates positional information on the caution-for-use term in the input text, and generates a detection result including the positional information. A communicating unit 11, by transmitting the detection result to a user terminal 3, causes the user terminal 3 to display, on a screen, the input text in a form in which a position of a term corresponding to the caution-for-use term is made apparent using the positional information.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a term detection device and a program for detecting terms that require attention when used from text.

Background Art

[0002] Conventionally, there is known a telop display technique for superimposing text related to broadcast content on an image of a television receiving a broadcast or displaying text such as emergency bulletins.

[0003] The telop display technique is widely used as a screen display technique, and there are also those that display opinions or the like arbitrarily input by users as telops on a screen, such as not only in the television broadcasting industry but also on video sites on the Internet.

[0004] As an example of such a telop display technique, a telop generation method that does not require a dedicated editing technique has been proposed for application to diversified video output techniques (see, for example, Patent Document 1).

[0005] This method generates a telop using various shared information such as information previously owned in the television broadcasting industry, information that can be provided by public institutions such as the government and local public bodies, and information publicly disclosed by general companies. Specifically, this method extracts information necessary for creating a telop from the shared information based on a telop creation request such as information regarding the distribution destination of the telop to be generated and information regarding the content of the telop display, and generates a telop using the extracted information.

[0006] However, there has been a problem that the telop generated by such a method may contain terms that require attention when used, and it is difficult to determine whether the generated telop can be used as it is.

[0007] To solve such problems, a method is envisioned to detect terms that require attention from the generated captions using a database (DB) in which terms that require attention are registered.

[0008] For example, although not targeted at captions, a method has been proposed to determine whether a text posted on an electronic bulletin board contains terms that require attention (see, for example, Patent Document 2).

[0009] This method determines whether a text posted on an electronic bulletin board contains terms that require attention. If such terms are included, the method sends to the user a web page data in which the text and the URL of the electronic bulletin board are listed so as to highlight the terms.

[0010] On the other hand, a method has been proposed to realize an easy-to-read caption display for users who receive text presentation and try to understand its content (see, for example, Patent Document 3).

[0011] This method replaces Chinese characters included in the text with kana according to the grade of the user for users in elementary and junior high schools.

[0012] Specifically, this method obtains the grade of the user, identifies Chinese characters that the user cannot read from the text based on the obtained grade of the user and an assignment grade table (a table in which Chinese characters and the grades for learning the Chinese characters are defined), and replaces the identified Chinese characters with kana based on ruby character data for caption display.

Prior Art Documents

Patent Documents

[0013]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0014] The method of detecting terms that require attention using the aforementioned DB determines whether the text contains terms that require attention by comparing the terms registered in the DB with the text of the generated caption.

[0015] However, this method has a problem that the capacity of the DB becomes large because it is necessary to register various variations of terms that require attention in the DB. There is also a problem that the processing load when comparing the terms registered in the DB with the text becomes high and the processing takes time.

[0016] For this reason, when detecting terms that require attention from text, it has been desired to reduce the load of the detection process. Also, when presenting the detection result to the user, it has been desired to present the terms that require attention in an easily understandable form.

[0017] Therefore, the present invention has been made to solve the above problems, and an object thereof is to provide a term detection device and a program that reduce the processing load when detecting terms that require attention from text and present the detection result to the user in an easily understandable form.

Means for Solving the Problems

[0018] To solve the above problems, the term detection device according to claim 1 receives, as input text, text input by a user from a user terminal, and detects a predetermined term from the input text. In the term detection device, a term that requires attention when used is defined as a term requiring attention, and assuming that a plurality of terms requiring attention are registered in advance as term-attention information, a storage unit storing the term-attention information, a morphological analysis unit that performs morphological analysis on the input text to generate text composed of a plurality of morphemes, a verb substitution unit that extracts a morpheme in the conjugated form of a verb from the text composed of the plurality of morphemes generated by the morphological analysis unit, replaces the conjugated form of the verb with its base form, and generates text after verb substitution, a detection unit that performs pattern matching between the text after verb substitution generated by the verb substitution unit and each of the plurality of terms requiring attention registered in the term-attention information stored in the storage unit, detects the term requiring attention from the text after verb substitution, and generates position information of a term corresponding to the detected term requiring attention from the input text, and the user terminal is caused to display the input text on a screen in a form in which the position of the term included in the input text is clearly indicated using the position information generated by the detection unit.

[0019] Further, the term detection device according to claim 2 is the term detection device according to claim 1, wherein the term-attention information registers, for each of the plurality of terms requiring attention, information indicating the degree of attention when using the term requiring attention, and the detection unit further reads, from the term-attention information stored in the storage unit, information indicating the degree of attention corresponding to the detected term requiring attention, and causes the user terminal to display the input text on a screen in a form in which the position of the term included in the input text is clearly indicated using the position information generated by the detection unit, and at the same time, displays, corresponding to the position of the term included in the input text, information indicating the degree of attention read by the detection unit on the screen.

[0020] Furthermore, the program according to claim 3 uses terms that require attention when in use as terms for attention, and assuming that a plurality of terms for attention are registered in advance as term-for-attention information, it includes a storage unit in which the term-for-attention information is stored, receives text input by a user as input text from a user terminal, and uses the term-for-attention information stored in the storage unit to detect terms corresponding to the terms for attention from the input text. A computer that constitutes a term detection device, a morphological analysis unit that performs morphological analysis on the input text to generate text consisting of a plurality of morphemes, extracts morphemes in the conjugated form of a verb from the text consisting of the plurality of morphemes generated by the morphological analysis unit, replaces the conjugated form of the verb with its base form, and generates text after verb replacement, and performs pattern matching between the text after verb replacement generated by the verb replacement unit and each of the plurality of terms for attention registered in the term-for-attention information stored in the storage unit, detects the terms for attention from the text after verb replacement, and functions as a detection unit that generates position information of terms corresponding to the detected terms for attention from the input text, and causes the user terminal to display the input text on the screen in a form in which the position of the term included in the input text is clearly indicated using the position information generated by the detection unit. This is the gist of the invention.

Advantages of the Invention

[0021] As described above, according to the present invention, when detecting terms that require attention from text, the processing load can be reduced, and the detection result can be presented to the user in an easy-to-understand form.

Brief Description of the Drawings

[0022]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0023] Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. Hereinafter, regarding the present invention, a system for supporting the work of creating subtitles (subtitle creation support system) will be taken as an example, and a process example of detecting terms that require attention from the text used for subtitles and displaying the detection results on the screen will be described.

[0024] Hereinafter, terms that require attention when used will be described as "terms for precautions". The "terms for precautions" include not only broadcast prohibited terms, broadcast caution terms, broadcast self-restraint terms, broadcast problem terms, etc. in the television broadcasting industry, but also terms prohibited from use in other industries, terms that require attention when used, etc. The "terms for precautions" is a general term for these.

[0025] In addition, it consists of a plurality of terms for precautions, etc., and for each term for precautions, a group of information including information related to the term for precautions will be described as "term information for precautions". The "term information for precautions" consists of related information such as reading, degree of attention (level), explanation, etc. for each term for precautions, and is pre-registered by the user (see FIG. 7 described later).

[0026] 〔Subtitle Creation Support System〕 First, a system including a subtitle creation support device (term detection device) according to an embodiment of the present invention will be described. FIG. 1 is a schematic diagram showing an overall configuration example of the subtitle creation support system.

[0027] This subtitle production support system 1 is a system that supports the task of producing subtitles for educational programs, and is configured to include a subtitle production support device (term detection device) 2 and a user terminal 3. The subtitle production support device 2 and the user terminal 3 are connected by a network 4 such as an intranet.

[0028] The user accesses a web page (a program for performing the processing of the subtitle production support system 1) stored in the subtitle production support device 2 by operating, for example, a browser operating on the user terminal 3. Then, with the user's access to the web page, the program is executed, and each function of the subtitle production support device 2 and the user terminal 3 described below is realized.

[0029] The subtitle production support device 2 receives the text input according to the user operation from the user terminal 3, detects the terms for precautions from the text, and transmits the detection result to the user terminal 3, so that the user terminal 3 displays the detection result on the screen together with the text. The details of the subtitle production support device 2 will be described later.

[0030] The user terminal 3 inputs text according to the user operation and transmits the text to the subtitle production support device 2. Then, the user terminal 3 receives the detection result for the text from the subtitle production support device 2 and displays the detection result on the screen together with the text. In addition, the user terminal 3 inputs a predetermined condition according to the user operation, and if there are Chinese characters that match the predetermined condition together with the input text, for example, the color of the characters at the corresponding location is changed. The details of the user terminal 3 and the predetermined condition will be described later.

[0031] 〔Subtitle Production Support Device 2〕 Next, the subtitle production support device 2 shown in FIG. 1 will be described in detail. FIG. 2 is a block diagram showing a configuration example of the subtitle production support device 2 according to an embodiment of the present invention.

[0032] As described above, the caption production support device 2 is a term detection device that receives the text input according to the user operation on the user terminal 3, detects the cautionary terms from the text, and transmits the detection result to the user terminal 3, thereby causing the user terminal 3 to display the detection result on the screen.

[0033] This caption production support device 2 includes a communication unit 11, a morphological analysis unit 12, a verb replacement unit 13, a detection unit 14, and a storage unit 15. The detection unit 14 includes a cautionary term detection means 21. In addition, a morphological analysis dictionary 31 and cautionary term information 32 are stored (registered) in advance in the storage unit 15.

[0034] In addition, the detection unit 14 may include a Chinese character detection means 22 in addition to the cautionary term detection means 21, and grade-specific Chinese character information 33 may be stored in advance in the storage unit 15 in addition to the morphological analysis dictionary 31 and the cautionary term information 32. Also, the user terminal 3 may include the Chinese character detection means 22, or the grade-specific Chinese character information 33 may be stored in advance in the storage unit of the user terminal 3. That is, the process of the Chinese character detection means 22 using the grade-specific Chinese character information 33 is performed by either the caption production support device 2 or the user terminal 3.

[0035] Since the morphological analysis dictionary 31 is known, detailed description thereof is omitted here. Details of the cautionary term information 32 will be described later. Also, the grade-specific Chinese character information 33 is a list of Chinese characters according to the Course of Study, and the Chinese characters (Chinese characters to be learned) to be learned according to the grade are defined. Since the grade-specific Chinese character information 33 is known, detailed description thereof is omitted here.

[0036] The communication unit 11 receives the text input according to the user operation from the user terminal 3, performs a predetermined reception process, and outputs the text after the reception process to the morphological analysis unit 12. Also, the communication unit 11 inputs the detection result from the detection unit 14, performs a predetermined transmission process, and transmits the detection result after the transmission process to the user terminal 3.

[0037] The morphological analysis unit 12 receives text from the communication unit 11 and performs morphological analysis on the text using the morphological analysis dictionary 31 stored in the storage unit 15. The morphological analysis unit 12 then generates text consisting of multiple morpheme terms classified by parts of speech, and outputs this to the verb replacement unit 13 as text after morphological analysis.

[0038] The verb replacement unit 13 receives the morpheme-analyzed text from the morpheme analysis unit 12 and extracts morpheme terms indicating the conjugated form of the verb from the morpheme-analyzed text. The verb replacement unit 13 then replaces the morpheme terms indicating the conjugated form of the verb with morpheme terms indicating the original form of the verb, generating text including the replaced original form of the verb. The verb replacement unit 13 outputs this to the detection unit 14 as the text after verb replacement.

[0039] For example, referring to FIG. 4(2)(3) described later, if the morpheme term indicating the conjugated form of a verb is "kurutsu", this term is replaced with the morpheme term "kuruu" indicating the original form of the verb.

[0040] Each verb has one root form and multiple conjugated forms. This replacement process is carried out using multiple sets of data (data consisting of root forms and multiple conjugated forms for all verbs) that are set in advance, with the root form of the verb and multiple corresponding conjugated forms as one set.

[0041] The caution-required term detection means 21 of the detection unit 14 receives the text after verb replacement from the verb replacement unit 13 .

[0042] The caution word detection means 21 performs pattern matching between the text after the verb replacement and each of a plurality of caution words included in the caution word information 32 stored in the storage unit 15. Then, the caution word detection means 21 detects one or more caution words that match in the pattern matching from the text after the verb replacement.

[0043] For example, if the text after verb replacement contains the term "kuuru" which is a morpheme indicating the base form of a verb, and the usage caution term information 32 contains the usage caution term for "kuuru", then the usage caution term "kuuru" is detected by pattern matching.

[0044] For each detected usage caution term, the usage caution term detection means 21 identifies the term corresponding to the detected usage caution term from the text (input text) output by the communication unit 11 to the morphological analysis unit 12, and generates position information indicating the position of that term included in the input text.

[0045] In the above example, the term "kurutta" corresponding to the detected usage caution term "kuuru" is identified, and the position information of that term "kurutta" included in the input text is generated. Note that as will be described later, since the term "kurutta" is an inflected form of a verb, the term "kurutta" including the auxiliary verb "ta" following that term is identified and the position information is generated.

[0046] The usage caution term detection means 21 reads out the related information (reading, degree of caution, explanation, etc.) corresponding to the detected usage caution term from the usage caution term information 32 stored in the storage unit 15. Then, the usage caution term detection means 21 generates a detection result including the position information and the related information for the detected usage caution term, and outputs the detection result to the communication unit 11.

[0047] When the detection unit 14 includes the Chinese character detection means 22, the Chinese character detection means 22 inputs the grade-specific detection conditions (not shown) input according to the user operation at the user terminal 3 via the communication unit 11. The details of the grade-specific detection conditions will be described later. Then, the Chinese character detection means 22 extracts one or more Chinese characters from the input text, and for each of the extracted Chinese characters, uses the grade-specific Chinese character information 33 stored in the storage unit 15 to identify the grade (the grade for learning that Chinese character) for learning guidance of that Chinese character according to the Course of Study.

[0048] The Chinese character detection means 22 determines, for each of the extracted Chinese characters, whether or not it matches the detection conditions for each grade, generates a detection result including the Chinese characters that match the detection conditions for each grade, and outputs the detection result to the communication unit 11.

[0049] Then, by the communication unit 11 transmitting the detection result generated by the usage caution term detection means 21 to the user terminal 3, the telop production support device 2 causes the user terminal 3 to display the input text on the screen in a form in which the position of the term corresponding to the usage caution term is clearly indicated using the position information included in the detection result. Further, the telop production support device 2 causes the user terminal 3 to display the relevant information on the screen according to the position where the term corresponding to the usage caution term included in the input text is displayed.

[0050] Also, by the communication unit 11 transmitting the detection result generated by the Chinese character detection means 22 to the user terminal 3, the telop production support device 2 causes the user terminal 3 to display the input text on the screen in a form in which it is clearly indicated that the Chinese characters that match the detection conditions for each grade match the detection conditions for each grade for the Chinese characters that match the detection conditions for each grade included in the detection result.

[0051] (Processing of the telop production support device 2) Next, the processing of the telop production support device 2 shown in FIG. 2 will be described in detail. FIG. 3 is a flowchart showing an example of the processing of the telop production support device 2 shown in FIG. 2, and FIG. 4 is a diagram for explaining an example of data from when the input text is received until the detection result is generated. FIG. 5 is a diagram for explaining an example of the input text, and FIG. 6 is a diagram for explaining an example of the detection result. FIGS. 3 to 6 show the case where the detection result of the usage caution term is generated by the usage caution term detection means 21 of the detection unit 14.

[0052] The communication unit 11 receives text (input text) from the user terminal 3 (step S301). Then, the morphological analysis unit 12 performs morphological analysis on the input text using the morphological analysis dictionary 31, and generates text (text after morphological analysis) composed of terms of a plurality of morphemes distinguished for each part of speech (step S302).

[0053] Referring to FIGS. 4(1), 4(2) and 5, for example, as shown in FIGS. 4(1) and 5, assume that the input text is "Pedaled the AEROBIKE (registered trademark) like crazy". Then, as shown in FIG. 4(2), the text after morphological analysis generated by the morphological analysis unit 12 is "ki" (noun) / "ga" (particle) / "kurutta" (conjugated form of verb) / "ta" (auxiliary verb) / "youni" (auxiliary verb) / "aerobike" (noun) / "wo" (particle) / "kogita" (conjugated form of verb) / "tsuzuketa" (conjugated form of verb) / "ta" (auxiliary verb).

[0054] In the input text shown in FIG. 5, "line": 0 (refer to o1) indicates that the line number of the first half part "ki ga kurutta youni" of the input text is 0. Also, "line": 1 (refer to o2) indicates that the line number of the second half part "aerobike wo kogitsuzuketa" of the input text is 1.

[0055] Returning to FIG. 3, the verb replacement unit 13 extracts the terms of the morphemes indicating the conjugated forms of the verbs from the text after morphological analysis generated by the morphological analysis unit 12, replaces these with the terms of the morphemes indicating the base forms of the verbs, and generates the text after verb replacement (step S303).

[0056] Referring to FIGS. 4(2) and 4(3), the verb replacement unit 13 extracts the terms of the morphemes indicating the conjugated forms of the verbs "kurutta" (conjugated form of verb) / "kogita" (conjugated form of verb) / "tsuzuketa" (conjugated form of verb) from the text after morphological analysis shown in FIG. 4(2).

[0057] Then, the terms of the morphemes indicating the conjugated forms of the verbs "kurutta" (conjugated form of verb) / "kogita" (conjugated form of verb) / "tsuzuketa" (conjugated form of verb) are replaced with the terms of the morphemes indicating the base forms of the verbs "kurau" (base form of verb) / "kogu" (base form of verb) / "tsuzukeru" (base form of verb), respectively.

[0058] Then, as shown in FIG. 4(3), the text after verb substitution generated by the verb substitution unit 13 is "ki" (noun) / "ga" (particle) / "kuruu" (base form of verb) / "ta" (auxiliary verb) / "youni" (auxiliary verb) / "aerobike" (noun) / "wo" (particle) / "kogu" (base form of verb) / "tsuzukeru" (base form of verb) / "ta" (auxiliary verb).

[0059] Returning to FIG. 3, the usage caution term detection means 21 of the detection unit 14 performs pattern matching between the text after verb substitution and each of the plurality of usage caution terms registered in the usage caution term information 32, and detects a usage caution term from the text after verb substitution (step S304).

[0060] FIG. 7 is a diagram showing an example of the data configuration of the usage caution term information 32 registered in advance. This usage caution term information 32 is composed of a plurality of sets of data with a usage caution term (phrase), ruby, type, description, synonyms, and category as a set.

[0061] A usage caution term is a term that requires attention when used. For example, "ki ga kuruu", "kuruu", "aerobike", and "gum tape" are registered. A usage caution term consists of a term of one morpheme or a term formed by combining a plurality of morphemes.

[0062] When the part of speech of a usage caution term is a verb, only the base form of the verb is registered as a usage caution term, and the conjugated form of the verb is not registered. For example, the base form of the verb "kuruu" is registered as a usage caution term, and the conjugated form "kutta" is not registered. This is the same when the base form of the verb is included in the usage caution term, such as "ki ga kuruu". For example, "ki ga kuruu" including the base form of the verb is registered as a usage caution term, and "ki ga kutta" including the conjugated form of the verb is not registered.

[0063] As a result, when the caution term is a verb, only the original form of the verb needs to be registered in caution term information 32, and there is no need to register the conjugated form of the verb. The same applies when the caution term includes a verb. This makes it possible to reduce the capacity of caution term information 32 and reduce the load of the pattern matching process in step S304 by caution term detection means 21.

[0064] Ruby is auxiliary information such as furigana and notes for the word with caution. For example, ruby "kigakurau" for the word with caution "kigoruu" (go crazy) and ruby "earobaiku" for the word with caution "aero bike" are registered.

[0065] The type indicates the type that reflects the degree of care that must be taken when using the caution term. For example, "with great care" is registered, indicating that the term must be used with great care, and "can be used" is registered, indicating that the term can be used.

[0066] The explanation is a sentence that expresses the content or meaning of the caution term so that the user can understand it. The synonym indicates whether or not there is a synonym for the caution term, for example, "There is no synonym" or "There is a synonym." If there is a synonym, that synonym is also registered. The category indicates the field, category, department, area, etc. in which the caution term is used, for example, "Disease / Disability," "Trademark," and "Promotion / Advertisement / Trademark" are registered.

[0067] With reference to Figures 4 (3) and 7, the caution word detection means 21 performs pattern matching between the text after verb replacement, namely "ki", "ga", "kuruu", "ta", "youni", "aero bike", "o", "kogu", "tsuzureru", and "ta" (parts of speech omitted), and each of the multiple caution words included in the caution word information 32.

[0068] Then, as shown in FIG. 4(4), the terms that require caution in use, such as “go crazy,” “go crazy,” and “exercise bike,” are detected.

[0069] Returning to FIG. 3, the usage caution term detection means 21 identifies terms corresponding to the usage caution terms detected in step S304 from the input text received by the communication unit 11 in step S301. Then, the usage caution term detection means 21 generates position information of the term in the input text (step S305). The position information is composed of the line number l of the input text, and the indexes of the start position s and the end position e of the term in the input text.

[0070] In addition, when the part of speech of the term in the input text is an inflected form of a verb, the usage caution term detection means 21 generates position information including the auxiliary verb or case particle term following the term.

[0071] Referring to FIGS. 4(1), (4), and (5), the terms "went crazy", "went crazy", and "aerobike" corresponding to the usage caution terms "go crazy", "go crazy", and "aerobike" are identified from the input text "Rowed the aerobike like crazy" by the usage caution term detection means 21.

[0072] In this case, when the detected usage caution term is "go crazy", the corresponding term is "went crazy", but since the auxiliary verb term "ta" follows the inflected form verb term "went crazy", the term corresponding to the usage caution term "go crazy" is "went crazy". Similarly, the term corresponding to the detected usage caution term "go crazy" is "went crazy".

[0073] Then, referring to FIGS. 4(6) and 5, for the term "went crazy" in the input text, position information consisting of line number l = 0, start position s = 0, and end position e = 4 is generated by the usage caution term detection means 21.

[0074] The position information consists of the line number l of the input text including the term corresponding to the usage caution term in the input text, the start position s of the term in the input text of the line number, and the end position e of the term in the input text of the line number.

[0075] As shown in Fig. 5, since the line number of the input text containing the term "crazy" is "line": 0, the line number l = 0. Also, since the starting position of the term "crazy" included in the input text is at the 0th position when counting from 0 for "crazy", the starting position s = 0. Further, since the ending position of that term is at the 4th position when counting from 0 for "ta", the ending position e = 4.

[0076] Similarly, by the usage caution term detection means 21, for the term "crazy" in the input text, position information consisting of line number l = 0, starting position s = 2, and ending position e = 4 is generated. Also, by the usage caution term detection means 21, for the term "aero bike" in the input text, position information consisting of line number l = 1, starting position s = 0, and ending position e = 5 is generated.

[0077] Returning to Fig. 3, the usage caution term detection means 21 reads out related information such as the ruby, type, etc. corresponding to the usage caution term detected in step S304 from the usage caution term information 32 (step S306).

[0078] Referring to Fig. 4(4) and Fig. 7, by the usage caution term detection means 21, from the usage caution term information 32, related information of the ruby "kigakuruu", type "pay full attention", explanation "Words such as 'crazy' or 'head crazy' are about the spirit...", synonym "There are no synonyms", and category "disease / disorder" corresponding to the detected usage caution term "crazy" is read out.

[0079] Similarly, by the usage caution term detection means 21, from the usage caution term information 32, related information such as the ruby "kuruu", type "pay full attention", etc. corresponding to the detected usage caution term "crazy" is read out. Also, by the usage caution term detection means 21, from the usage caution term information 32, related information such as the ruby "earobaiiku", type "pay full attention", etc. corresponding to the detected usage caution term "aero bike" is read out.

[0080] Returning to FIG. 3, for each of the usage caution terms detected in step S304, the usage caution term detection means 21 generates, as detection results, the position information generated in step S305 and the related information such as the ruby, type, etc. read in step S306 (step S307).

[0081] Referring to FIG. 6, the usage caution term detection means 21 generates, as detection results, the related information and position information corresponding to the detected usage caution term "going crazy" ("phrase": "going crazy", see p1).

[0082] The related information in this case is the ruby "kigakuruu" ("ruby": "kigakuruu", see p2), the type "be extremely careful" ("type": "be extremely careful", see p3), the description "Words such as 'going crazy' or 'losing one's mind' refer to the spirit..." ("description": "Words such as 'going crazy' or 'losing one's mind' refer to the spirit...", see p4), the synonyms "There are no synonyms" ("synonyms": "There are no synonyms", see p5), and the category "disease / disorder" ("category": "disease / disorder", see p6).

[0083] Also, the position information is the line number l = 0, the start position s = 0, and the end position e = 4 ("position": {"line": 0, "start": 0, "end": 4}, see p7).

[0084] Similarly, the usage caution term detection means 21 generates, as detection results, the related information such as the ruby "kuruu" ("ruby": "kuruu") and the type "be extremely careful" ("type": "be extremely careful") corresponding to the detected usage caution term "crazy" ("phrase": "crazy"), and the position information consisting of the line number l = 0, the start position s = 2, and the end position e = 4 ("position": {"line": 0, "start": 2, "end": 4}).

[0085] Also, the usage caution term detection means 21 generates, as a detection result, related information such as a ruby "earobaiku" ("ruby": "earobaiku"), a type "be extremely careful" ("type": "be extremely careful"), etc. corresponding to the detected usage caution term "aerobike" ("phrase": "aerobike"), and position information consisting of line number l = 1, start position s = 0, and end position e = 5 ("position": {"line": 1, "start": 0, "end": 5}).

[0086] Returning to FIG. 3, the communication unit 11 transmits the detection result generated by the usage caution term detection means 21 in step S307 to the user terminal 3 (step S308).

[0087] That is, the communication unit 11 transmits the detection result to the user terminal 3 so that the input text is displayed on the screen in a form in which the positions of the terms (the terms indicated by the position information, the terms corresponding to the detected usage caution terms) included in the input text are specified using the position information included in the detection result.

[0088] Also, the communication unit 11 transmits the detection result to the user terminal 3 so that the related information is displayed on the screen according to the position of the term corresponding to the usage caution term included in the input text using the related information included in the detection result.

[0089] As a result, the user terminal 3 receives, from the telop production support device 2, a detection result consisting of related information and position information about the usage caution terms as a response to the input text.

[0090] The user terminal 3 extracts the position information and related information of each usage caution term from the detection result. Then, based on the position information, the user terminal 3 specifies the position of the term (the term corresponding to the usage caution term) included in the input text, and displays the input text on the screen in a form in which the position is specified (for example, highlighted display, display in a specific color).

[0091] Furthermore, the user terminal 3 displays related information on the screen in accordance with the position of the term corresponding to the term with caution. For example, the user terminal 3 displays related information, such as the term with caution, ruby, type, explanation, synonym, and category, on the screen near the position of the term corresponding to the term with caution and so as not to overlap with the input text.

[0092] 8 is a diagram showing an example of a display screen of the user terminal 3, showing an example when supporting the task of producing subtitles for an educational program. This example of the display screen shows that the input text is "He kept pedaling the exercise bike like he was going crazy" (see the text input area h2). It also shows that the detected terms that should be used with caution are "going crazy", "going crazy", and "exercise bike", and that the terms that correspond to the terms that should be used with caution included in the input text are "went crazy", "went crazy", and "exercise bike" (see i1, i2 in the result display area h3).

[0093] This example display screen also shows that for the input text "Kiki ni natte yo doke kaze wo aerobaikko wo kaze wo kimi ga tsukare" (I kept pedaling the exercise bike like crazy), if the kanji learned in the 4th grade and above are converted to hiragana, the kanji that match are "Kyou" (mad) and "Zoku" (continue) (see j1 and j2 in the result display area h3). Furthermore, it shows that if the kanji learned from the 2nd to 3rd grade are converted to kanji with ruby notation, there are no kanji that match.

[0094] Also, on the left side of the example display screen, the detection conditions by grade level are displayed (see the condition input area h1). The detection conditions by grade level are conditions that reflect the production rules of educational programs, for example, when producing subtitles for educational programs.

[0095] In this example display screen, the detection conditions by grade are that kanji learned in the 4th grade and above should be converted to hiragana, and that kanji learned from the 2nd to 3rd grade should be converted to ruby. Other items displayed include an item for specifying whether to align all alphanumeric characters and symbols in the input text to full-width or half-width, and an item for specifying whether to save the detection results in text format.

[0096] First, the user inputs the detection conditions by grade by operating the keyboard and mouse using the user terminal 3. Specifically, when the user wants to detect Chinese characters by grade, in the condition input area h1, the user specifies to turn on the switch of "detect Chinese characters by grade", and specifies the grades of the Chinese characters with ruby characters and the Chinese characters to be converted into hiragana characters.

[0097] In this example of the display screen, as the detection conditions by grade, it is specified that the Chinese characters learned by fourth graders and above are to be converted into hiragana characters, and the Chinese characters learned from second graders to third graders are to be ruby characters.

[0098] In addition, when the user wants to detect the terms for precautions, the user specifies to turn on the switch of "detect the terms for precautions" (not shown in the figure). Then, the user inputs the text to be detected (input text) in the text input area h2.

[0099] Then, when the user terminal 3 is equipped with the Chinese character detection means 22 and the grade-specific Chinese character information 33 shown in FIG. 2, the Chinese character detection means 22 extracts Chinese characters from the input text according to the input text input according to the user operation. Then, for each of the extracted Chinese characters, the Chinese character detection means 22 uses the grade-specific Chinese character information 33 to identify the grade for receiving learning guidance, determines whether it meets the grade-specific detection conditions, and generates a detection result including the Chinese characters that meet the grade-specific detection conditions. In the case of this example, a detection result is generated in which the Chinese characters learned by fourth graders and above are to be converted into hiragana characters, and the Chinese characters learned from second graders to third graders are to be ruby characters.

[0100] The user terminal 3 (the display unit) displays the input text on the screen in the result display area h3, and also displays the kanji included in the input text that is the detection result in a form that corresponds to the "kanji to add ruby" and the "kanji to be converted to hiragana". In the example display screen of FIG. 8, since "ki" is a kanji learned in the first grade, and "kyo" and "zuku" are kanji learned in the fourth grade and above, "kyo" and "zuku" are displayed on the screen in a predetermined form as the kanji to be converted to hiragana (see j1, j2 in the result display area h3). For example, the color of the characters "kyo" and "zuku" is changed.

[0101] Furthermore, when input text is entered in the text input area h2 by a user operation, for example, when the Enter key is pressed, the user terminal 3 transmits the input text to the subtitle production support device 2. In this case, the user terminal 3 transmits newly entered input text to the subtitle production support device 2 every time the Enter key is pressed.

[0102] As a result, the subtitle production support device 2 performs a process of detecting words that should be used with caution from the input text, and the detection result is transmitted from the subtitle production support device 2 to the user terminal 3. For example, the detection result shown in Fig. 6 is transmitted.

[0103] The user terminal 3 (display unit) displays the input text on the screen in the result display area h3 based on the detection results sent from the subtitle production support device 2, and highlights the terms "crazy," "crazy," and "exercise bike," which correspond to the cautionary terms, in the input text displayed on the screen (see i1, i2).

[0104] Here, the user terminal 3 (the display unit) moves the cursor to the position of the term "went crazy" in accordance with the user's operation. Then, the user terminal 3 (the display unit) displays on the screen related information (ruby, type, explanation, etc.) about the usage caution term "went crazy", which is the original form of the verb for the term "went crazy", near the position of the term "went crazy" and so as not to overlap with the input text (see k).

[0105] In this way, the user can recognize the cautionary terms included in the text of the input text, and regarding the cautionary terms, can recognize the content such as the reading method, the degree of caution, its meaning, etc., the presence or absence of synonyms, and the category. Also, the user can recognize the Chinese characters for each school year according to the curriculum guidelines as the characters with ruby or the characters to be written in hiragana.

[0106] Thereby, for example, when the user creates a telop to be used in a broadcast program, by using the telop production support system 1, it is possible to check the content to be displayed on the telop at the work stage instead of the preview stage. For this reason, it is possible to reduce the rework of the telop production work at the preview stage.

[0107] As described above, according to the telop production support apparatus 2 of the embodiment of the present invention, the morphological analysis unit 12 performs morphological analysis on the input text received from the user terminal 3, and the verb replacement unit 13 extracts the term of the morpheme indicating the conjugated form of the verb from the text after morphological analysis, and replaces this with the term of the morpheme indicating the base form of the verb.

[0108] The cautionary term detection means 21 of the detection unit 14 performs pattern matching between the text after verb replacement and each of the plurality of cautionary terms registered in the cautionary term information 32, and detects the cautionary term from the text after verb replacement. Then, the cautionary term detection means 21 identifies the term corresponding to the detected cautionary term from the input text, and generates position information indicating the position of the term in the input text.

[0109] Also, the cautionary term detection means 21 reads out the related information (reading method, degree of caution, explanation, etc.) corresponding to the detected cautionary term from the cautionary term information 32, and generates a detection result including the position information and the related information.

[0110] The communication unit 11 transmits the detection result to the user terminal 3, causing the user terminal 3 to display the input text on the screen in a form where the position of the term corresponding to the usage caution term is indicated using the position information included in the detection result, and to display the related information included in the detection result on the screen according to the position of the term corresponding to the usage caution term included in the input text.

[0111] As a result, the user can instantly recognize at which position in the input text the detected usage caution term is included. Also, even when the input text includes the conjugated form (conjugated verb) of a verb, the user can recognize the usage caution term.

[0112] In addition, since the pattern matching by the usage caution term detection means 21 is performed only on the base form of the verb, it is sufficient that only the base form of the verb is registered in the usage caution term information 32, and there is no need to register various variations of the conjugated form. For this reason, the capacity of the usage caution term information 32 can be reduced, and the processing load when detecting the usage caution term from the input text can be reduced.

[0113] Also, the telop production support device 2 causes the user terminal 3 to display the input text on the screen in a form where the position of the term corresponding to the usage caution term is indicated, and to display the related information on the screen according to the position of the term corresponding to the usage caution term included in the input text. As a result, the usage caution term can be presented to the user in an easily understandable form.

[0114] In this case, the user can distinguish between usability and non-usability and understand the degree of caution required when using the usage caution term by recognizing, for example, the type (such as "fully caution", "usable", etc.) included in the related information of the usage caution term displayed on the screen.

[0115] In addition, in the embodiments of the present invention, when a user creates a caption for a broadcast program, since it is possible to detect terms for usage precautions at the work stage rather than at the preview stage, it is possible to reduce the rework of the caption production work at the preview stage.

[0116] As described above, the present invention has been described with reference to the embodiments. However, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the technical idea thereof.

[0117] For example, the caption production support system 1 shown in FIG. 1 is used to support the work of producing captions. However, the present invention is applicable not only to the caption production support system 1, but also to, for example, a system that supports the work of producing scripts for broadcast programs and the like. In short, the present invention is applicable to various systems that detect terms that require attention from the text to be detected and present the detection results to the user.

[0118] In addition, in the caption production support system 1 shown in FIG. 1, the user terminal 3 is configured to input detection conditions according to the school year that reflect the production rules when producing captions for educational programs according to user operations. In this case, the user terminal 3 or the caption production support device 2 generates a detection result including Chinese characters that match the detection conditions according to the school year from the input text, and the user terminal 3 displays the detection result on the screen.

[0119] On the other hand, the user terminal 3 may be configured to input conditions that reflect the production rules when producing captions according to the type of program other than educational programs. In this case, the caption production support device 2 or the user terminal 3 generates a detection result including terms that satisfy this condition from the input text, and the user terminal 3 displays the detection result on the screen.

[0120] As for the hardware configurations of the caption production support device 2 and the user terminal 3 according to the embodiments of the present invention, a normal computer can be used. The caption production support device 2 and the user terminal 3 are configured by a computer including a CPU, a volatile storage medium such as a RAM, a non-volatile storage medium such as a ROM, and an interface or the like.

[0121] Each function of the communication unit 11, the morphological analysis unit 12, the verb replacement unit 13, the detection unit 14, and the storage unit 15 provided in the caption production support device 2 is realized by causing the CPU to execute a program describing these functions. The same applies to each function such as the display unit provided in the user terminal 3.

Explanation of Signs

[0122] 1 Caption production support system 2 Caption production support device 3 User terminal 4 Network 11 Communication unit 12 Morphological analysis unit 13 Verb replacement unit 14 Detection unit 15 Storage unit 21 Cautionary term detection means 22 Chinese character detection means 31 Morphological analysis dictionary 32 Cautionary term information 33 Chinese character information by school year l Line number s Start position e End position h1 Condition input area h2 Text input area h3 Result display area

Claims

1. In a term detection device that receives text input by a user as input text from a user terminal and detects a predetermined term from the input text, assuming that terms that require attention when used are defined as terms requiring attention, and a plurality of terms requiring attention are registered in advance as term attention information, a storage unit storing the term attention information, a morphological analysis unit that performs morphological analysis on the input text to generate text consisting of a plurality of morphemes, a verb replacement unit that extracts a morpheme in the conjugated form of a verb from the text consisting of the plurality of morphemes generated by the morphological analysis unit, replaces the conjugated form of the verb with its base form, and generates text after verb replacement, performs pattern matching between the text after verb replacement generated by the verb replacement unit and each of the plurality of terms requiring attention registered in the term attention information stored in the storage unit, detects the term requiring attention from the text after verb replacement, and a detection unit that generates position information of a term corresponding to the detected term requiring attention from the input text, and characterized in that the input text is displayed on the screen of the user terminal in a form in which the position of the term included in the input text is clearly indicated using the position information generated by the detection unit.

2. In the term detection device according to Claim 1, information indicating the degree of attention when using each of the plurality of terms requiring attention is registered in the term attention information, the detection unit further reads out information indicating the degree of attention corresponding to the detected term requiring attention from the term attention information stored in the storage unit, the input text is displayed on the screen of the user terminal in a form in which the position of the term included in the input text is clearly indicated using the position information generated by the detection unit, and information indicating the degree of attention read out by the detection unit is displayed on the screen corresponding to the position of the term included in the input text.

3. A term detection device that detects terms that require caution when used and that include a plurality of terms that require caution when used and that are registered in advance as term information for terms that require caution, the device comprises a storage unit that stores the term information for terms that require caution when used, receives text input by a user from a user terminal as input text, and detects terms that correspond to the terms that require caution when used from the input text using the term information for terms that require caution when used stored in the storage unit, a morpheme analysis unit that performs morpheme analysis on the input text and generates a text consisting of a plurality of morphemes; a verb replacement unit that extracts morphemes of verb conjugations from the text including the plurality of morphemes generated by the morpheme analysis unit, replaces the verb conjugations with their original forms, and generates a verb-replaced text; and performing pattern matching between the verb-replaced text generated by the verb replacement unit and each of the plurality of caution words registered in the caution word information stored in the storage unit, and detecting the caution words from the verb-replaced text; a detection unit that generates, from the input text, position information of a term corresponding to the detected careful usage term; A program characterized by causing the user terminal to display the input text on a screen in a form in which the position of the term contained in the input text is clearly indicated, using the position information generated by the detection unit.

Images