Search device, search method, and program

Abstract

To efficiently search for a target example sentence from a plurality of example sentences by using emotion information.SOLUTION: A search device comprises: a receiving unit which receives designation of a search condition including a search word and emotion information; a search unit which searches for an example sentence which includes the search word and with which the designated emotion information is associated, from a plurality of example sentences and emotion analysis information associated with the plurality of example sentences; and an output control unit which causes an output device to output an example sentence being a search result of the search unit.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The disclosure of this specification relates to a search device, a search method, and a program. [Background technology]

[0002] There is known an electronic dictionary technology that allows users to search for example sentences from dictionary data. Such a technology is described in, for example, Patent Document 1. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-134567 Summary of the Invention [Problem to be solved by the invention]

[0004] However, while conventional techniques can search for example sentences, if the search results contain a large number of example sentences, the user may not be able to find the example sentence they are looking for. In light of the above-described circumstances, an object of one aspect of the present invention is to provide a technique for efficiently searching for a desired example sentence from a plurality of example sentences by utilizing emotion information. [Means for solving the problem]

[0005] A search device according to one aspect of the present invention includes a reception unit that receives a specified search condition including a search word and emotional information, a search unit that searches a plurality of example sentences and emotional analysis information linked to the plurality of example sentences for an example sentence that includes the search word and is associated with the specified emotional information, and an output control unit that outputs the example sentences that are the search results of the search unit to an output device. [Effects of the Invention]

[0006] According to the above aspect, by utilizing emotion information, it is possible to efficiently search for a desired example sentence from among a plurality of example sentences. [Brief explanation of the drawings]

[0007] [Figure 1] 1 is a diagram illustrating a configuration of a system including a search device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating an example of a functional configuration of a search device. [Figure 3] FIG. 10 is a diagram illustrating an example of a home screen displayed on the output device. [Figure 4] FIG. 10 is a diagram illustrating an example of a search result screen displayed on an output device. [Figure 5] FIG. 10 is a diagram illustrating an example of a narrowing down selection screen displayed on the output device. [Figure 6] FIG. 10 is a diagram illustrating an example of an emotion selection screen displayed on the output device. [Figure 7] FIG. 10 is a diagram illustrating an example of a narrowing-down screen displayed on the output device. [Figure 8] 10 is a flowchart illustrating an example of an example sentence search process. [Figure 9] FIG. 10 is a diagram illustrating an example of a functional configuration of a search device according to another embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a search result screen displayed on an output device according to another embodiment. [Figure 11] FIG. 10 is a diagram illustrating an example of a selection operation screen displayed on an output device according to another embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of an emotion selection screen displayed on an output device according to another embodiment. [Figure 13] FIG. 10 is a diagram illustrating an example of a selection analysis screen displayed on an output device according to another embodiment. [Figure 14] 10 is a flowchart illustrating an example of an example sentence search process according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] 1 is an example of an electronic dictionary that accesses content such as electronic dictionary data and question collection data, searches for example sentences from the content, and displays the example sentences as search results. This search device 100 may be configured as a search system 1 that includes a server device 200 that is accessed via a network N such as the Internet.

[0009] The search device 100 is a terminal that is directly operated by a user. The search device 100 may be any terminal, such as a mobile terminal, notebook PC, tablet, or smartphone, as long as it can run an application program that searches for and displays example sentences of content.

[0010] 1, the search device 100 includes a processor 110, a storage device 120, a communication device 130, an input device 140, and an output device 150. The processor 110 is, for example, a central processing unit (CPU) or a graphics processing unit (GPU). When the processor 110 executes a program 121 stored in the storage device 120, the computer that is the search device 100 operates as a reception unit 111, a search unit 112, an output control unit 113, and an information generation unit 114 shown in FIG. 2 to perform various processes.

[0011] The storage device 120 includes, for example, semiconductor memory that operates as a main storage device, such as a RAM (Random Access Memory) or a ROM (Read Only Memory), and storage that operates as an auxiliary storage device, such as an SSD (Solid State Drive) or an HDD (Hard Disk Drive). The storage device 120 is an example of a storage device. The storage device 120 stores a program 121 executed by the processor 110 and various databases (example sentence information database 122) used when the program 121 is executed.

[0012] The example sentence information database 122 contains sentiment analysis information for content that includes at least example sentences. The content includes electronic dictionary data, question collection data, etc. The electronic dictionary data is a collection of dictionary data related to various dictionaries, such as an English-Japanese dictionary, a German-Japanese dictionary, a French-Japanese dictionary, and a Japanese dictionary. The electronic dictionary data includes at least headword words and example sentence collections containing example sentences corresponding to the headwords, in a corresponding state. The electronic dictionary data is an example of dictionary content. Searches can be performed based on example sentences included in the example sentence collections of the dictionary content, thereby providing support for users when searching in the dictionary. The question collection data is a collection of question collection data related to various questions, such as fill-in-the-blank questions and multiple-choice questions. The question collection data includes example sentences that constitute at least part of the questions. In this embodiment, example sentences refer to sentences or passages given as examples or illustrations. The question collection data is an example of question collection content. Example sentences may be composed of at least one word, and include short sentences, complex sentences, compound sentences, etc.

[0013] The sentiment analysis information is information in which multiple example sentences constituting the content are associated with emotional information acquired by emotionally analyzing the multiple example sentences constituting the content, and is stored in the example sentence information database 122. Therefore, the sentiment analysis information is configured by associating each example sentence with emotional information corresponding to each example sentence. The sentiment analysis for acquiring the emotional information is performed in advance using AI (artificial intelligence) such as machine learning, by inputting example sentences included in the example sentence collection of the content of the electronic dictionary data or example sentences included in the question sentences of the content of the question collection data, analyzing the emotions of the example sentences, and outputting the emotional information associated with the example sentences. The content of the electronic dictionary data is an example of dictionary content. The sentiment analysis information is stored so that one example sentence is associated with one piece of emotional information. Since the sentiment analysis information is stored in advance in the storage device 120, the search speed by the search device 100 can be improved. Note that multiple pieces of emotional information may be stored in the sentiment analysis information so that one example sentence is associated with multiple pieces of emotional information. When sentiment analysis information is stored in which multiple pieces of emotional information are associated with one example sentence, a score value may be assigned to each piece of emotional information. This allows for more accurate searches based on the score values. Furthermore, the emotion analysis information may include an emotion key associated with each example sentence as emotion information. When an emotion key is associated with each example sentence, the search unit 112 can efficiently search for and narrow down the example sentences to the user's desired example sentence from multiple example sentences based on the emotion key.

[0014] Emotions refer to the feelings a user has toward an object or thing, and emotional information includes at least one of joy, anger, sadness, happiness, negative, and positive emotions. Joy, anger, sadness, and happiness include joy, anger, sadness, and pleasure, respectively. For example, joy refers to feelings such as happiness and satisfaction, anger refers to feelings that arise in response to dissatisfaction or injustice, sadness refers to feelings such as sadness, disappointment, and loneliness, and pleasure refers to feelings that arise from comfort, a pleasant state, or a pleasant experience. Negative emotions refer to feelings that weigh heavily on people's hearts, such as anxiety, sadness, and anger, while positive emotions refer to feelings that brighten people's hearts, such as joy, gratitude, and hope. Users can search for example sentences that correspond to emotions from multiple example sentences based on emotional information that expresses various emotions, so by using emotional information, they can efficiently search for a desired example sentence from multiple example sentences.

[0015] The communication device 130 is a communication module that communicates with the server device 200 via the network N. The input device 140 accepts user operations. The input device 140 is a keyboard, a touch panel, a mouse, etc. The output device 150 outputs information to the user. The output device 150 is a liquid crystal display, an organic EL display, etc. If the output device 150 is a liquid crystal display, an organic EL display, etc., the output device displays images of example sentences, etc. to the user. The input device 140 and the output device 150 may be formed integrally. If the output device 150 is a speaker, etc., the output device outputs audio of example sentences, etc. to the user.

[0016] The search device 100 configured as described above is configured to provide a service for searching for example sentences in content in response to a request from the input device 140, and also to display example sentences from dictionaries and workbooks within the content. This point will be explained in more detail below.

[0017] Below, the functional configuration related to the example sentence search processing of the search device 100 will be described with reference to Fig. 2, and the screen transitions displayed by the search device 100 will be described with reference to Figs. 3 to 7. As shown in Fig. 2, the search device 100 includes a reception unit 111, a search unit 112, and an output control unit 113, and the example sentence search processing is realized by the processor 110 executing a program 121.

[0018] When the example sentence search process starts, the processor 110 displays a home screen 310 shown in FIG. 3 on the output device 150. The home screen 310 includes an input field 311 for inputting a search word to be searched for based on a user's input operation. The accepting unit 111 accepts, based on a user's input operation on the input device 140, a search condition including at least one search word and emotion information. After accepting the search word as a search condition, the accepting unit 111 accepts emotion information as a search condition for narrowing down example sentences containing the search word. Therefore, after a search using a search word, example sentences can be narrowed down based on emotion information. This allows the number of example sentences to be processed to be reduced using the search word in a first stage, and then narrowed down based on emotion information in a second stage, thereby improving processing speed. Furthermore, if there are few example sentences in the search results when an example sentence search based on a search word is performed and the desired example sentence can be found, it is possible to avoid narrowing down the search results using emotion information, thereby improving the convenience of the search device 100. Specifically, when an electronic dictionary is selected as content in the search device 100, the reception unit 111 receives, as a search word, a word that the user searches for as a search target via the input device 140. As shown in Fig. 3, when "take" is entered as a word to be searched for in an input field 311 based on an input operation by the user via the input device 140, the reception unit 111 receives the entered word "take" as a search word. The reception unit 111 may receive the designation of emotion information as a search criterion for narrowing down example sentences, and then may receive the designation of a search word as a search criterion, or may simultaneously receive the designation of a search word as a search criterion and emotion information as a search criterion for narrowing down example sentences containing the search word.

[0019] The search unit 112 searches the emotion analysis information stored in the example sentence information database 122 for example sentences that include the search word specified by the reception unit 111 and that are associated with emotion information specified by the reception unit 111. The search unit 112 searches the content included in the example sentence information database 122 for example sentences that include the search word, based on the search word specified as a search condition received by the reception unit 111.

[0020] When the search unit 112 searches for multiple example sentences containing the search word, the output control unit 113 displays the search result screen 320 shown in FIG. 4 on the output device 150 as search results, including the multiple example sentences 321 to 329 searched by the search unit 112. The search result screen 320 displays the multiple example sentences 321 to 329 searched by the search unit 112 and a refine / word order button 330. The multiple example sentences 321 to 329 are example sentences that include the word "take" accepted by the acceptance unit 111 as a search word. Although many example sentences can be searched for based on the search word in this manner, if many example sentences are displayed as search results, the user may not be able to find the example sentence he or she is looking for. Therefore, in this embodiment, the search unit 112 searches for example sentences that include the search word and are associated with specified emotion information from the emotion analysis information stored in the example sentence information database 122. As a result, the emotion information can be used to search for and display the example sentence he or she is looking for from multiple example sentences. As a result, by utilizing emotion information, it is possible to narrow down and display the example sentences that the user desires from among a plurality of example sentences.

[0021] The receiving unit 111 determines whether or not a selection for narrowing down the search results has been received. The receiving unit 111 can determine whether or not a selection for narrowing down the search results has been received based on whether or not a selection of the narrowing down / word order button 330 has been received, based on an input operation by the user. When a selection for narrowing down the search results has been received, that is, when a selection of the narrowing down / word order button 330 has been received, the output control unit 113 displays a narrowing down selection screen 340 shown in FIG. 5 on the output device 150. The narrowing down selection screen 340 is a screen that, when a plurality of example sentences searched by the search unit 112 are displayed on the output device 150 as search results, receives a selection of a method for narrowing down the displayed plurality of example sentences, based on a user operation. The narrowing down selection screen 340 displays buttons 341 to 344 that receive a selection of a method for narrowing down the displayed plurality of example sentences. Of buttons 341 to 344 that accept the selection of a method for narrowing down the multiple example sentences, button 341 accepts the selection of a method for displaying candidates corresponding to emotions, and each of buttons 342 to 344 accepts the selection of a method for displaying candidates based on the word order of the entered words or the consecutive use of entered words. This allows a wide variety of methods for narrowing down the multiple displayed example sentences, allowing the user to narrow down the example sentences in the way they desire.

[0022] The reception unit 111 determines whether a method of displaying candidates corresponding to emotions has been selected, based on a user's selection operation on the button 341. If the reception unit 111 determines that a method of displaying candidates corresponding to emotions has been selected, the output control unit 113 displays an emotion selection screen 350 shown in FIG. 6 on the output device 150. The emotion selection screen 350 is an example of an emotion filter that searches for example sentences by emotion. The emotion selection screen 350 displays a plurality of emotion selection buttons 351 to 354 that allow the user to select each emotion. The emotion selection screen 350 is an example of the reception unit 111 that receives the specification of search conditions based on emotion information.

[0023] Each emotion selection button 351 to 354 corresponds to a desired emotion of the user. When each emotion selection button 351 to 354 is selected based on an input operation by the user, the reception unit 111 receives a search condition specification based on the emotion of joy, anger, sadness, or enjoyment, respectively.

[0024] The receiving unit 111 determines, based on the user's input operation, which emotion has been selected based on the user's input operation for any of the emotion selection buttons 351 to 354. In this embodiment, it is assumed that a search condition based on the emotion of joy is specified as a search condition based on emotion information. That is, when the emotion selection button 351 corresponding to the emotion of joy is selected based on the user's input operation, the receiving unit 111 receives the specification of a search condition based on emotion information of joy corresponding to the emotion selection button 351.

[0025] When the search condition based on the emotion information is specified, the search unit 112 searches the emotion analysis information stored in the example sentence information database 122 for example sentences that include the search word and are associated with the specified emotion information.

[0026] When the search unit 112 has searched for example sentences associated with emotion information, the output control unit 113 displays a narrowing-down screen 360 shown in Fig. 7 on the output device 150 as the search results for the example sentences associated with emotion information by the search unit 112. The narrowing-down screen 360 displays example sentences 322, 325, 326, 327, and 329 as search results. The example sentences 322, 325, 326, 327, and 329 are obtained by searching for example sentences 322, 325, 326, 327, and 329 associated with emotion information from among the multiple example sentences 321 to 329 searched for by the search unit 112. The narrowing-down screen 360 shown in Fig. 7 displays example sentences 322, 325, 326, 327, and 329 searched for based on a search condition based on emotion information of joy. That is, example sentences 322, 325, 326, 327, and 329 are examples of search results obtained by searching, based on search criteria based on the emotion of joy, among multiple example sentences 321 to 329 that contain the search word searched for by search unit 112. In this way, even if there are multiple example sentences that contain the search word, example sentences can be searched for and narrowed down based on search criteria based on emotion, and the user can easily search for a desired example sentence from multiple example sentences by using emotion information.

[0027] In the search device 100, the processor 110 performs the example sentence search process shown in Fig. 8 to search for and display example sentences that include the search word and that are associated with specified emotion information from the emotion analysis information stored in the example sentence information database 122. Specifically, first, the processor 110 displays the home screen 310 shown in Fig. 3 on the output device 150 (step S1), and determines whether or not the search word has been accepted (step S2). The processor 110 waits for the process to finish until the search word is accepted (step S2: NO). Upon accepting the search word (step S2: YES), the processor 110 searches the emotion analysis information stored in the example sentence information database 122 for example sentences that include the search word as a search condition (step S3), and displays the search result screen 320 shown in Fig. 4 on the output device 150 as the search results for the plurality of example sentences 321 to 329 (step S4). Processor 110 determines whether a narrowing-down selection has been accepted (step S5), and waits until a narrowing-down selection has been accepted (step S5: NO). If a narrowing-down selection has been accepted (step S5: YES), processor 110 displays narrowing-down selection screen 340 shown in Fig. 5 on output device 150 (step S6). Processor 110 determines whether a method of displaying candidates corresponding to emotions has been selected (step S7), and waits until a method of displaying candidates corresponding to emotions has been selected (step S7: NO). If a method of displaying candidates corresponding to emotions has been selected (step S7: YES), processor 110 displays emotion selection screen 350 shown in Fig. 6 on output device 150 (step S8). Processor 110 determines whether a search condition based on emotion information has been accepted (step S9). Processor 110 waits until it receives the specification of search conditions based on emotion information (step S9: NO), and when it receives the specification of search conditions based on emotion information (step S9: YES), it searches the emotion analysis information stored in example sentence information database 122 for example sentences 321 to 329 that include the search word "take" and that are associated with "joy" as the specified emotion information (step S10).Processor 110 displays narrowing screen 360 shown in Fig. 7 on output device 150 as the search results, which are example sentences 322, 325, 326, 327, and 329 associated with emotion information through the search (step S11). When this process ends, the example sentence search process ends.

[0028] As described above, many example sentences 321 to 329 containing search words such as "take" are displayed, which may prevent the user from finding the example sentence they are looking for. In contrast, in this embodiment, example sentences that contain the search word and are associated with specified emotion information can be searched for and displayed based on emotion analysis information stored in the example sentence information database 122. This makes it possible to narrow down and display the example sentence that the user is looking for from multiple example sentences by using emotion information, thereby making it easy for the user to search for the example sentence they are looking for.

[0029] In the above-described embodiment, the sentiment analysis information is stored in the storage device 120 in association with a plurality of example sentences constituting the content and emotion information acquired by sentiment analysis of the plurality of example sentences constituting the content, but this is not limited thereto. As shown in FIG. 2 , the search device 100 may further include an information generation unit 114 implemented by the processor 110 executing a program 121. The information generation unit 114 analyzes and generates emotion analysis information in real time based on the plurality of example sentences and emotion information acquired by sentiment analysis of the plurality of example sentences. When the search device 100 includes the information generation unit 114, the search unit 112 may search for example sentences that include the search word specified by the reception unit 111 and are associated with the emotion information specified by the reception unit 111, from the emotion analysis information generated by the information generation unit 114 through analysis in real time. This eliminates the need to previously have the sentiment analysis information learned by AI and stored in the storage device 120, thereby reducing the storage capacity of the storage device 120.

[0030] In the above-described embodiment, the search device 100 implements the program 121 by the processor 110, but this is not limited to this. The server device 200 may execute the program to execute the example sentence search process and display the example sentences on the search device. In this case, the processor of the server device 200 can be provided with a reception unit, a search unit, an output control unit, and an information generation unit corresponding to the reception unit 111, the search unit 112, the output control unit 113, and the information generation unit 114 included in the processor 110. This allows the search device 100 to process the example sentences found by the server device 200 simply by displaying them on the output device, thereby reducing the processing power.

[0031] In the above-described embodiment, the processor 110 searches for example sentences that include a search word and are associated with specified emotion information from the emotion analysis information stored in the example sentence information database 122. However, for example, as shown in FIG. 9 , the processor 110 may perform emotion analysis on a selected portion of an example sentence displayed on the output device 150 for which a selection operation has been accepted, output emotion information for the selected portion for which the selection operation has been accepted, and display the output emotion information on the output device 150. Specifically, as shown in FIG. 9 , the processor 110 further includes an emotion output unit 115. The emotion output unit 115 performs emotion analysis on a selected portion 422 of an example sentence 421 displayed on the output device 150 for which a user selection operation has been accepted, and outputs emotion information for the selected portion 422 for which the user selection operation has been accepted. The emotion analysis for outputting the emotion information is performed in real time using AI such as machine learning.

[0032] When the sentiment analysis process starts, the processor 110 displays a home screen 310 shown in Fig. 3 on the output device 150. The search unit 112 searches for example sentences containing the search word specified by the reception unit 111 from the sentiment analysis information stored in the example sentence information database 122. The home screen 310 shown in Fig. 3 is the same as that in the above-described embodiment, and therefore will not be described again.

[0033] When the search unit 112 searches for an example sentence containing a search word, the output control unit 113 displays a search result screen 420 shown in FIG. 10 on the output device 150 as a search result, which is an example sentence 431 containing the search word searched by the search unit 112. The search result screen 420 displays the example sentence 431 searched for by the search unit 112. The example sentence 431 is an example sentence containing the word "take" accepted by the acceptance unit 111 as a search word. Although example sentences can be searched for based on search words in this way, it may be difficult for a user to understand the emotion of the example sentence. Therefore, in another embodiment, the emotion output unit 115 performs emotion analysis on a selected portion of the example sentence for which a selection operation has been accepted, outputs emotion information for the selected portion, and displays the emotion information on the output device 150. As a result, the emotion of the selected portion of the example sentence can be analyzed to support the user's reading comprehension of the sentence.

[0034] The receiving unit 111 determines whether or not a selection operation has been received from among the portions constituting the example sentence 431 displayed on the output device 150. The selection operation refers to an operation of receiving a highlighting or touching operation from among the words, phrases, etc. constituting the example sentence 431 based on an input operation by the user. When a selection operation has been received, the output control unit 113 displays a selection operation screen 430 shown in FIG. 11 on the output device 150. On the selection operation screen 430, a selected portion 422, for which a selection operation has been received, from among the portions constituting the example sentence 431 displayed on the output device 150 is displayed in a highlighted state. In addition, an analysis button 423 is displayed on the selection operation screen 430.

[0035] When the analysis button 423 is selected based on a user operation, the emotion output unit 115 performs an emotion analysis on the selected portion 422 for which the selection operation has been accepted, and outputs emotion information for the selected portion 422 for which the selection operation has been accepted. The emotion information is output by inputting an example sentence of the selected portion 422, and calculating (deriving) the emotion feature amount of the example sentence of the selected portion 422, and using AI such as machine learning to output at least one emotion parameter corresponding to the example sentence of the selected portion 422. The emotion parameter is calculated so as to increase in accordance with the calculated emotion feature amount.

[0036] When the emotion output unit 115 calculates the emotion feature amount and outputs the emotion information, the output control unit 113 displays an emotion selection screen 440 shown in FIG. 12 on the output device 150. The emotion selection screen 440 displays the emotion information output by the emotion output unit 115 using parameters 441. In FIG. 12, parameters 441 corresponding to emotions such as joy, anger, sadness, fun, and fear are output as emotion information. The emotion information is not limited to that shown in FIG. 12, and parameters corresponding to various emotions can be output. Each parameter is displayed larger according to the emotion feature amount. By looking at the parameters 441 corresponding to each emotion, the user can easily understand the emotion contained in the example sentence in the selected portion 422. This can support the user's reading comprehension of the example sentence in the selected portion 422.

[0037] The receiving unit 111 determines whether an emotion corresponding to the emotion information displayed by the emotion output unit 115 on the output device 150 has been selected. In FIG. 12 , emotion information 442 of joy has been selected based on the user's input operation. As a result of the determination, the emotion portion corresponding to the emotion information received by the receiving unit 111 is displayed in a manner different from the others. Specifically, the output control unit 113 displays a selection analysis screen 450 shown in FIG. 13 on the output device 150. The output control unit 113 displays emotion portions 443 and 444 having high feature amounts of emotion information 442 selected for the selected portion 422 in a manner different from the others. The term "display different from the others" means any manner that allows the user to recognize that the emotion portions 443 and 444 corresponding to the emotion information 442 received by the receiving unit 111 are different from other words, and examples of such a display include emphasis, highlighting, inversion, blinking, bold line display, and dotted line display. By displaying the emotional part 443 corresponding to the emotional information 442 in a different manner from the others, it is possible to understand which part of the example sentence corresponds to the emotion with a high feature value of the emotional information 442, and it is possible to support the user in reading and comprehending the example sentence.

[0038] In the search device 100, the processor 110 performs the sentiment analysis process shown in Fig. 14 to perform sentiment analysis on a selected portion of an example sentence that includes a search word and for which a selection operation has been accepted, based on sentiment analysis information stored in the example sentence information database 122, output sentiment information for the selected portion for which a selection operation has been accepted, and display the output sentiment information on the output device 150. Specifically, the processor 110 first displays the home screen 310 shown in Fig. 3 on the output device 150 (step S21), and determines whether or not a search word has been accepted (step S22). The processor 110 waits until the search word is accepted (step S22: NO). Upon accepting the search word (step S22: YES), the processor 110 searches for example sentences 431 that include the search word from the sentiment analysis information stored in the example sentence information database 122 (step S23), and displays the search result screen 420 shown in Fig. 10 on the output device 150 as the search result of the searched example sentence 431 (step S24). Processor 110 determines whether selection portion 422 has been selected (step S25) and waits until selection portion 422 has been selected (step S25: NO). If selection portion 422 has been selected (step S25: YES), processor 110 displays selection operation screen 430 shown in FIG. 11 on output device 150 and analyzes the emotion of selection portion 422 (step S26). Processor 110 determines whether analysis button 423 has been selected (step S27). Processor 110 waits until analysis button 423 has been selected (step S27: NO). If analysis button 423 has been selected, processor 110 calculates the emotion feature quantities of the example sentence in selection portion 422 (step S28). After the emotion feature quantities have been calculated and emotion information has been output, processor 110 displays emotion selection screen 440 shown in FIG. 12 on output device 150 and displays the output emotion information as parameters 441 (step S29). Processor 110 determines whether an emotion for emotional information has been selected (step S30), and waits until an emotion for emotional information has been selected (step S30: NO), and when an emotion for emotional information has been selected (step S30: YES), processor 110 highlights and displays emotional parts 443, 444 with high feature amounts of selected emotional information 442 for selected part 422 (step S31).When this process is completed, the sentiment analysis process is completed. The sentiment analysis process shown in Fig. 14 may be executed separately from or in parallel with the example sentence search process shown in Fig. 8.

[0039] As described above, in another embodiment, emotion analysis is performed on a selected portion of an example sentence that has been selected, emotion information for the selected portion 422 that has been selected is output, and the emotion information can be displayed on the output device 150. As a result, the emotion of the example sentence in the selected portion is analyzed and the output emotion information is displayed as parameters 441. By looking at the parameters 441 corresponding to each emotion, the user can easily understand the emotion of the example sentence in the selected portion 422. This can support the user's reading comprehension of the example sentence in the selected portion 422. Furthermore, when an emotion for the emotion information is selected, emotion portions 443 and 444 with high feature amounts of emotion information 442 selected for the selected portion 422 can be displayed in a different manner from the others. As a result, by looking at the emotion portions 443 and 444, it can be easily understood which parts of the example sentence in the selected portion 422 correspond to the emotion. This can support the user's reading comprehension of the example sentence in the selected portion 422.

[0040] The above-described embodiments are illustrative examples provided to facilitate understanding of the invention. The present invention is not limited to the above-described embodiments, and should be understood to encompass various modifications and alternative forms of the above-described embodiments. For example, it will be understood that the above-described embodiments can be embodied by modifying the components without departing from the spirit of the invention. It will also be understood that various embodiments can be implemented by appropriately combining multiple components disclosed in the above-described embodiments. Furthermore, it will be understood by those skilled in the art that various embodiments can be implemented by deleting some components from all of the components shown in the embodiments, or by adding some components to the components shown in the embodiments.

[0041] In the above-described embodiment, the search unit 112 accepts only one word as a search condition, but this is not limited to this. For example, when two or more search words are specified as search conditions, the search unit 112 may search for example sentences that include all of the search words regardless of the order in which the words were input, example sentences that include all of the search words in the order in which the words were input, or example sentences that include all of the search words consecutively in the order in which the words were input, from the content included in the example sentence information database 122. This allows example sentences to be searched for in a wide variety of ways, allowing the user to search for example sentences in a way that suits their needs.

[0042] If the output device 150 is a speaker, the output control unit 113 may output aloud from the speaker the multiple example sentences searched by the search unit 112 as search results. In this case, in the search device 100, the processor 110 performs the example sentence search process shown in Fig. 8 to search for example sentences that include the search word and are associated with specified emotional information from the emotion analysis information stored in the example sentence information database 122, and output the example sentences aloud from the speaker. This allows the user to check the example sentences through audio, and by using the emotional information, the user can easily search for a desired example sentence from multiple example sentences. [Explanation of symbols]

[0043] 100: Search device, 111: Reception unit, 112: Search unit, 113: Output control unit

Claims

1. a reception unit that receives a search condition including a search word and emotion information; a search unit that searches for example sentences that include the search word and are associated with the specified emotion information from a plurality of example sentences and emotion analysis information linked to the plurality of example sentences; an output control unit that outputs the example sentences that are searched by the search unit to an output device; A search device characterized by:

2. 2. The search device according to claim 1, Further comprising a storage device for storing information; The emotion analysis information is information in which the example sentences and the emotion information acquired by performing emotion analysis on the example sentences are associated with each other, and the information is stored in advance in the storage device. A search device characterized by:

3. 2. The search device according to claim 1, an information generation unit that generates the emotion analysis information based on the plurality of example sentences and the emotion information acquired by performing emotion analysis on the plurality of example sentences. A search device characterized by:

4. 4. The search device according to claim 2, wherein: The receiving unit receives the input of the search word, and then receives designation of the emotion information as the search condition for narrowing down example sentences containing the search word. A search device characterized by:

5. 4. The search device according to claim 3, the information generation unit performs a sentiment analysis on a selected portion that has received a selection operation from among portions that constitute the example sentences output to the output device; an emotion output unit that outputs the emotion information for the selected portion that has received the selection operation, The output control unit outputs the emotion information output by the emotion output unit to the output device. A search device characterized by:

6. 6. The search device according to claim 5, the receiving unit receives a designation for the emotion information output by the emotion output unit to the output device; The output control unit outputs an emotion part corresponding to the emotion information received by the receiving unit in a manner different from others. A search device characterized by:

7. The computer Accepting a search condition including a search word and emotion information; searching for example sentences that include the search word and are associated with the specified emotion information from a plurality of example sentences and emotion analysis information linked to the plurality of example sentences; and outputting the example sentences as search results to an output device. A search method characterized by:

8. On the computer, A process of accepting a search condition including a search word and emotion information; a process of searching for an example sentence that includes the search word and is associated with the specified emotion information from a plurality of example sentences and emotion analysis information linked to the plurality of example sentences; and outputting the example sentences as search results to an output device. A program characterized by:

Images