Demonstration device

Abstract

The present invention aims to provide a presentation system that effectively displays keywords for selecting the next slide in the presentation. The present invention provides a display device that includes a voice recognition unit 53, a conversation-derived term extraction unit 55, a search keyword storage unit 57, a search keyword extraction unit 59, a data storage unit 61, a matching page information extraction unit 63, a selection term extraction unit 65, and a selection term display unit 71.

Description

[0001] This application is a divisional application of an application with a filing date of May 7, 2018, application number 201880035833.8, and invention name “Display Device”. Technical Field

[0002] The present invention relates to a display device for conversation support, which can suggest a suitable presentation page corresponding to the conversation during the conversation. Background Art

[0003] For example, Japanese Patent No. 4551105 discloses a conference support system using voice recognition. Voice recognition recognizes voice as text, so it is impossible to read a suitable slide. In addition, even if an attempt is made to read a slide using only the terms recognized by voice recognition, the terms may not be consistent with the search terms associated with the slide, so it is impossible to read a suitable slide.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent No. 4551105 Summary of the invention

[0007] Problems to be solved by the invention

[0008] The present invention aims to provide a presentation system which can effectively display keywords for selecting the next slide in the presentation.

[0009] Technical solutions to problems

[0010] The present invention is basically based on the knowledge that not only the terms included in the conversation but also the terms related to the terms are used to extract page candidates of presentation materials related to the conversation, so that page candidates of presentation materials related to the conversation can be appropriately suggested.

[0011] The present invention relates to a display device including a computer or a mobile terminal device, which can display information of a page for reading a presentation material associated with a conversation.

[0012] This display device includes a voice recognition unit 53 , a conversation-derived word extraction unit 55 , a search keyword storage unit 57 , a search keyword extraction unit 59 , a data storage unit 61 , a corresponding page information extraction unit 63 , a selected word extraction unit 65 , and a selected word display unit 71 .

[0013] The voice recognition unit 53 is a component for performing voice recognition.

[0014] The conversation-derived term extraction unit 55 is for extracting a plurality of conversation-derived term elements from the conversation information recognized by the voice recognition unit 53 .

[0015] The search keyword storage unit 57 is a component for storing conversation-derived words and search keywords in association with each other.

[0016] The search keyword extraction unit 59 is a component for extracting a plurality of search keywords from the search keyword storage unit 57 using the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit 55 .

[0017] The document storage unit 61 is a component for storing each page of a plurality of presentation documents in association with a search term for each page and a score indicating the probability of becoming the search term for each page.

[0018] The corresponding page information extraction unit 63 is a unit for extracting a page of a presentation material associated with the search keyword from the material storage unit 61 using the search keyword extracted by the search keyword extraction unit 59 as a search term.

[0019] The selection term extraction unit 65 is for extracting a search keyword as a selection term for selecting a slide when the presentation material page extracted by the matching page information extraction unit 63 exists.

[0020] The selected term display unit 71 is a component for displaying the selected term extracted by the selected term extraction unit 65 on the display unit 69 .

[0021] The display device has the above-mentioned various means, so that the selected words can be displayed on the display unit 69.

[0022] In the display device,

[0023] The matching page information extracting unit 63 may use the search keyword extracted by the search keyword extracting unit 59 as a search term and extract one or more pages of presentation materials associated with the search keyword from the material storage unit 61 for the search terms with higher scores.

[0024] In the display device,

[0025] The display unit 69 is a display screen of the terminal device.

[0026] The selection term display unit 71 displays the selection term extracted by the selection term extraction unit 65 in a selection term display area 73 located at the lower part of the display screen.

[0027] The display device may further include:

[0028] The selection word input unit 75 receives information that the displayed selection word is selected by the user and a specific selection word has been selected; and

[0029] The candidate page reading unit 77 uses the selection word inputted by the selection word input unit 75 to read page candidates of a plurality of presentation materials associated with the selection word.

[0030] Next, the display device may further include a page selection information input unit 79 for receiving information on a page of a specific presentation material selected from among the plurality of page candidates of the presentation material read by the candidate page reading unit 77 .

[0031] At this time, the display device may use the information related to the page of the presentation material selected by the page selection information input unit 79 and display the selected page of the presentation material on the display unit 69 .

[0032] The display device preferably further comprises a retrieval material information storage device 23, wherein the retrieval material information storage device 23 comprises:

[0033] A term extraction unit 3 extracts terms from the document, i.e., terms contained in a certain page of the document;

[0034] A keyword storage unit 5 stores terms that are keywords associated with terms in the material;

[0035] The keyword extraction unit 7 extracts a plurality of keywords associated with the terms in the document from the keyword storage unit 5 using the terms in the document extracted by the term extraction unit 3;

[0036] A subject word storage unit 9 stores subject words associated with keywords;

[0037] The subject word extraction unit 11 uses the multiple keywords extracted by the keyword extraction unit 7 to extract subject words associated with the keywords from the subject word storage unit 9;

[0038] The search term candidate extraction unit 13 extracts the search term candidates for a certain page of the document from the subject words extracted by the subject word extraction unit 11 and the plurality of keywords extracted by the keyword extraction unit 7;

[0039] The search term candidate display unit 17 displays the search term candidates extracted by the search term candidate extraction unit 13 on the display unit 15;

[0040] a search term input unit 19 for receiving an input indicating a search term among the search term candidates displayed on the display unit 15; and

[0041] The document search information storage unit 21 associates the search term input by the search term input unit 19 with information related to a certain page of the document and stores the information.

[0042] The present invention also provides a program for causing a computer to function as a display device, or a computer-readable recording medium having the program recorded thereon;

[0043] The display device has the following means, so that the selected words can be displayed on the display unit 69;

[0044] A voice recognition unit 53 for performing voice recognition;

[0045] The conversation-derived word extraction unit 55 extracts a plurality of conversation-derived words from the conversation information recognized by the voice recognition unit 53;

[0046] The search keyword storage unit 57 associates and stores the conversation-derived words with the search keywords;

[0047] The search keyword extraction unit 59 extracts a plurality of search keywords from the search keyword storage unit 57 using the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit 55;

[0048] The document storage unit 61 associates and stores each page of the plurality of presentation documents with a search term for each page and a score indicating the ease of becoming the search term for each page;

[0049] The matching page information extraction unit 63 uses the search keyword extracted by the search keyword extraction unit 59 as a search term, and extracts a page of the presentation material associated with the search keyword from the material storage unit 61;

[0050] The selection term extraction unit 65 extracts the search keyword as a selection term for selecting a slide when the presentation material page extracted by the matching page information extraction unit 63 exists; and

[0051] The selected term display unit 71 is for displaying the selected term extracted by the selected term extraction unit 65 on the display unit 69 .

[0052] Effects of the Invention

[0053] The present invention provides a presentation system which effectively displays keywords for selecting the next slide in the presentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a block diagram for explaining the display device of the present invention.

[0055] Figure 2 It is a block diagram showing the basic structure of a computer.

[0056] Figure 3 This is a conceptual diagram showing an example of a system of the present invention.

[0057] Figure 4 This is an example of a page of presentation material.

[0058] Figure 5 This is a conceptual diagram showing a storage example of the keyword storage unit.

[0059] Figure 6 This is a conceptual diagram showing a storage example of a keyword storage unit.

[0060] Figure 7 This is a conceptual diagram showing a storage example of the classifier storage unit.

[0061] Figure 8 It is a concept map showing the extracted (classification terms), subject terms, keywords, and terms used in the data.

[0062] Fig. 9 This is an example of a display screen.

[0063] Fig.10 This is a flowchart for explaining an example of using the search material information storage device of the present invention.

[0064] Fig.11 This is a conceptual diagram for explaining an example of using the search material information storage device of the present invention.

[0065] Fig.12 It is a block diagram for explaining the retrieval material information storage device of the present invention.

[0066] Fig.13 This is a flowchart for explaining an example of a process of displaying a selected term or a process of displaying a presentation material page associated with the selected term.

[0067] Fig.14 This is a conceptual diagram showing a display example of the display unit before the selection term is displayed.

[0068] Fig.15 This is a conceptual diagram showing a display example of the display unit when displaying a selection term.

[0069] Fig.16 This is a conceptual diagram showing a state where candidate pages are displayed on the display unit. DETAILED DESCRIPTION

[0070] The following describes specific embodiments of the present invention using the drawings. The present invention is not limited to the embodiments described below, and also includes those appropriately modified within the scope of knowledge by those skilled in the art from the following embodiments.

[0071] The present invention relates to a display device including a computer or a mobile terminal device, which can display information for reading a presentation material page associated with a conversation.

[0072] Figure 1This is a block diagram for explaining the display device of the present invention. This device is a processing device based on a computer. The computer can be any one of a mobile terminal device, a desktop personal computer and a server, or a combination of two or more. These are usually connected in a manner that can send and receive information via the Internet (intranet) or the like. It is also possible to make any computer have a part of the functions, etc., and use multiple computers to share the functions.

[0073] Figure 2 It is a block diagram showing the basic structure of a computer. As shown in this figure, the computer has an input unit 31, an output unit 33, a control unit 35, a computing unit 37 and a storage unit 39, and each element is connected by a bus 41 and the like and is configured in a manner that can send and receive information. For example, the storage unit can store a control program or various information. When predetermined information is input from the input unit, the control unit reads the control program stored in the storage unit. Then, the control unit appropriately reads the information stored in the storage unit and transmits it to the computing unit. In addition, the control unit transmits appropriate input information to the computing unit. The computing unit uses the various information received to perform computing processing and stores it in the storage unit. The control unit reads the computing results stored in the storage unit and outputs them from the output unit. Various processing is performed in this manner. The means for performing the various processing are referred to as various means.

[0074] Figure 3 1 is a conceptual diagram showing an example of a system of the present invention. Figure 3 As shown, the system of the present invention (system including the apparatus of the present invention) may also be a system including a mobile terminal device 45 connected to the Internet or an intranet 43, and a server 47 connected to the Internet or the intranet 43. Of course, a single computer or mobile terminal device may also function as the apparatus of the present invention, and a plurality of servers may also exist.

[0075] The monitor or display of the mobile terminal device 45 can function as the display unit 69 .

[0076] like Figure 1 As shown, this display device (device of the present invention) has: a voice recognition unit 53, a conversational term extraction unit 55, a search keyword storage unit 57, a search keyword extraction unit 59, a data storage unit 61, a matching page information extraction unit 63, a selected term extraction unit 65 and a selected term display unit 71.

[0077] The sound recognition unit 53 is a component for performing sound recognition. Sound recognition is a well-known technology. In order to perform sound recognition, it has a microphone that generally collects sound, and a sound analysis unit that analyzes the sound input from the microphone. In addition to Japanese Patent Gazette No. 4551105 (Patent Document 1), the sound recognition technology itself is a well-known technology as described in Japanese Patent Gazette No. 6127422, Japanese Patent Gazette No. 6114210 and Japanese Patent Gazette No. 6107003. Through the sound recognition unit 53, the sound contained in the conversation can be input into the device as input information, and the words contained in the conversation can be stored as data. The sound recognition unit 53 appropriately stores the words contained in the conversation in the storage unit. In addition, at this time, the sound recognition unit 53 can store the relevant information of the volume or volume change together with the words while storing the words.

[0078] The conversation-derived term extraction unit 55 is used to extract a plurality of conversation-derived terms from the conversation information recognized by the sound recognition unit 53. As described above, the terms included in the conversation are appropriately stored in the storage unit. In addition, the storage unit stores terms that can be used for retrieval. This term can be, for example, a noun. In addition, for example, the storage unit can have a term extraction database that stores terms used to extract the terms included in the conversation in response to medical purposes, banking purposes, etc. In this case, the conversation-derived term extraction unit 55 only needs to refer to the term extraction database to read the term extraction term, and determine whether the term extraction term is consistent with the conversation-derived term stored in the storage unit. Then, when the term extraction term has a conversation-derived term that is consistent with the term extraction term, the term is stored in the storage unit as a "conversation-derived term". At this time, the frequency of occurrence, volume, or volume change of the "conversation-derived term" in the conversation during a predetermined period can also be stored in the storage unit. In addition, the database for term extraction may also store scores of the ease with which individual terms can be represented as search terms together with the terms used for term extraction. In this case, the conversational term extraction unit 55 may extract candidates for "conversational terms" using one or more of the scores, frequencies, and volume changes as indicators, such as "conversational terms" with higher scores, "conversational terms" with higher frequencies, and "conversational terms" with changed volume (increased volume). In addition, for each of the above-mentioned requirements, for example, coefficients may be stored in advance, and various coefficients may be multiplied to obtain the scores of conversational terms. For example, when the coefficient is 1, there is no change in the score. On the other hand, when the volume increases significantly, the coefficient may be assigned to 1.1 to 1.9 according to the degree of change, thereby adjusting the score of conversational terms.

[0079] The search keyword storage unit 57 is a component for associating and storing conversation-derived terms with search keywords. Conversation-derived terms are basically terms that are assumed to be included in a conversation (e.g., a conversation between a business person and a customer). On the other hand, there may be inconsistencies between terms that are assumed to be included in a conversation and terms that are suitable for retrieval. The difference between spoken and written terms is one of them. Therefore, in the present invention, the search keyword storage unit 57 associates and stores conversation-derived terms with search keywords, and converts them into search keywords that are suitable for retrieval.

[0080] The search keyword storage unit 57 can be implemented by a storage unit (storage device) or by a database and a database management system. The storage device stores search keywords corresponding to individual conversational terms for a plurality of conversational terms. The storage device can also record scores associated with the search keywords and further used for search.

[0081] Examples of words derived from conversational terms include “diet weight,” “slim,” and “slender,” and examples of search keywords corresponding thereto include “slim body,” “obese,” “body shape,” and “physical examination.”

[0082] The search keyword extraction unit 59 is a unit for extracting a plurality of search keywords from the search keyword storage unit 57 using the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit 55. The search keyword extraction unit 59 extracts the search keywords stored in the search keyword storage unit 57 in association with the conversation-derived terms for each of the plurality of conversation-derived terms. At this time, the scores stored together with the search keywords may also be extracted. Next, the search keyword extraction unit 59 stores the extracted search keywords (and scores) in the storage unit. Thereafter, for all (or a predetermined number of) conversation-derived terms, the search keywords (and scores) are extracted from the search keyword storage unit 57 and stored in the storage unit. The search keyword extraction unit 59 extracts the search keywords using the search keywords (and the scores) temporarily stored in the storage unit. At this time, the search keyword extraction unit 59 may also extract the search keywords with higher scores. The search keyword extraction unit 59 stores the extracted search keywords in the storage unit. At this time, the search keyword extraction unit 59 may store a plurality of search keywords and their individual scores together.

[0083] The document storage unit 61 is a component for storing each page of a plurality of presentation documents in association with a search term for each page and a score indicating the probability of becoming the search term for each page.

[0084] The corresponding page information extraction unit 63 is a component for using the search keyword extracted by the search keyword extraction unit 59 as a search term and extracting the presentation material page associated with the search keyword from the material storage unit 61. The material storage unit 61 stores each page of the presentation material associated with the search term (for example, together with a score or a ranking).

[0085] The corresponding page information extraction unit 63 uses the search keyword to obtain information for extracting each page of the presentation material having a high score or ranking from the material storage unit 61 .

[0086] The matching page information extracting unit 63 may use the search keyword extracted by the search keyword extracting unit 59 as a search term and extract one or more pages of presentation materials associated with the search keyword from the material storage unit 61 for the search terms with higher scores.

[0087] The selection term extraction unit 65 is used to extract the search keyword as the requirement for selecting the selection term for the slide when there is a page that matches the presentation material extracted by the page information extraction unit 63. The selection term can be one or more. In addition, the search term can be a predetermined number, such as three, four, five or six. The number of search terms can be determined using the size of the selection term display area in the display unit, the text size of the selection term, and the number of characters in the selection term. At this time, for example, the selection term extraction unit 65 obtains information about the length of the selection term display area. Next, the selection term extraction unit 65 obtains relevant information about the text size of the part where the selection term is displayed in the display unit. The selection term extraction unit 65 obtains relevant information about the size other than each text part when the selection term is displayed. After that, the selection term extraction unit 65 extracts the selection terms with higher priority one by one, for example, because of the reason of higher score or ranking, and when the selection terms are displayed in the case of one, two, three, four, five, or six selection terms, for example, the length of the selection term portion displayed on the display unit is calculated, and the length of the selection term portion is compared with the length of the portion where the selection terms are displayed. Then, when the length of the selection term portion is shorter than the displayed portion, one selection term is added. In this way, the selection term extraction unit 65 can extract an appropriate number of selection terms.

[0088] The selected term display unit 71 is a component for displaying the selected term extracted by the selected term extraction unit 65 on the display unit 69 .

[0089] The display device has the above-mentioned various means, so that the selected words can be displayed on the display unit 69.

[0090] Fig.13 This is a flowchart for explaining an example of a process of displaying a selected term or a process of displaying a page of presentation material associated with the selected term. S represents a process.

[0091] The person in charge holding the display device talks with the customer (conversation start: S01).

[0092] At this time, the sound recognition unit 53 performs sound recognition on the conversation related information, including the words included in the conversation (sound recognition process: S02). The conversation related information (including the words) after sound recognition is appropriately stored in the storage unit. In addition, the frequency of the words in a predetermined time, the increase or decrease of the volume when a certain word is uttered, the change of the sound wavelength, and other coefficients corresponding to the sound change can be pre-stored in the display device, and when analyzing the relevant information of the conversation, the predetermined coefficient associated with the word is read.

[0093] The conversation-derived term extraction unit 55 appropriately reads the conversation-related information recognized by the sound recognition unit 53 from the storage unit, and uses the read information to extract a plurality of conversation-derived terms (conversation-derived term extraction process: S03). The display device has a database related to conversation-derived terms, and by referring to this database, terms such as nouns that are effective for retrieval can be selected from the terms included in the conversation. In this way, conversation-derived terms are extracted. In addition, regarding conversation-derived terms, scores of conversation-derived terms associated with the conversation-derived terms can also be pre-stored in consideration of the possibility of use in retrieval.

[0094] The search keyword storage unit 57 associates and stores the conversation-derived words with the search keywords. At this time, the search keywords may be associated and stored with the scores of the search keywords in consideration of the possibility of being used in the search. The search keyword extraction unit 59 uses the multiple conversation-derived words extracted by the conversation-derived word extraction unit 55 (appropriately read from the storage unit) and extracts multiple search keywords from the search keyword storage unit 57 (search keyword extraction process: S04). In this way, instead of using the conversation-derived words (there are also cases where the conversation-derived words are directly used as search keywords), they are converted into search keywords and searched, so that the page of the appropriate presentation material can be searched.

[0095] The material storage unit 61 associates and stores each page of a plurality of presentation materials, the search terms of each page, and the scores indicating the ease of becoming the search terms of each page. Next, the matching page information extraction unit 63 uses the search keywords extracted by the search keyword extraction unit 59 as the search terms, and extracts the pages of the presentation materials associated with the search keywords from the material storage unit 61 (page extraction process: S05). The pages of the presentation materials do not need to be displayed in this process. When there are one or more associated pages of the presentation materials, information for reading the associated pages (such as information related to the material ID and the number of pages) can be extracted. The information for reading the extracted associated pages is appropriately stored in the storage unit. The matching page information extraction unit 63 can also use the search keywords extracted by the search keyword extraction unit 59 as the search terms, and extract one or more pages of the presentation materials associated with the search keywords from the material storage unit 61 with the search terms having higher scores.

[0096] Next, the selection term extraction unit 65 determines whether there is a page that matches the presentation material extracted by the page information extraction unit 63. If there is, the search keyword is extracted as a selection term for selecting a slide (selection term extraction step: S06). Alternatively, the search keyword may be directly extracted as a selection term. Next, the extracted selection term is appropriately stored in the storage unit.

[0097] The selected term display unit 71 displays the selected term extracted by the selected term extraction unit 65 (read from the storage unit as appropriate) on the display unit 69 (selected term display step: S07 ). Fig.14 : is a conceptual diagram showing a display example of the display unit before displaying the selection term. Fig.15 7 is a conceptual diagram showing an example of a display unit displaying a selected term. In this example, the selected term is displayed in the selected term display area 73 present at the lower part of the display screen. Fig.15 In the example of , the display unit is a touch screen type, and a plurality of (five in this example) selection terms are displayed in the selection term display area 73. In this way, the selection terms are displayed on the display unit.

[0098] For example, the person in charge (user) touches any one of the selection terms displayed on the touch screen (or, in the case where the display unit is a monitor of a desktop computer, selects the selection term while moving the cursor with a mouse or the like). In this way, the selection term input unit 75 inputs the information of the selected specific selection term in the display device, and the display device receives the input information (selection information input process: S08). The relevant information of the input specific selection term is appropriately stored in the storage unit.

[0099] In this way, the candidate page reading unit 77 uses the selected word input by the selected word input unit 75 (appropriately read from the storage unit) and reads information related to the page candidates of the plurality of presentation materials associated with the selected word. Using the read information, a reduced screen of the page candidates is created and displayed on the display unit (candidate page display step: S09). Fig.16 is a conceptual diagram showing a state where candidate pages are displayed on a display unit. Fig.16 In the example shown, when there are multiple page candidates of the read presentation materials, multiple page candidates of the presentation materials in a reduced size may be arranged and displayed at the bottom of the selection word display area 73, or the selection word display area 73 may be moved upward and displayed at the bottom thereof. Alternatively, this candidate page display process may be skipped.

[0100] When the display unit is a touch screen, the person in charge (user) presses the upper part of the iconized page candidate. In this way, the display device will know that a specific candidate has been selected. That is, the page selection information input unit 79 receives information that a specific presentation material page has been selected from the multiple page candidates of the presentation material read by the candidate page reading unit 77. Then, the received information is input into the display device. The input information (information that a specific page has been selected, or related information of the specified page) is appropriately stored in the storage unit. At this time, the display device uses the related information of the page of the presentation material selected by the page selection information input unit 79 (appropriately read from the storage unit), and displays the selected page of the presentation material on the display unit 69 (page display process: S10).

[0101] The display device preferably further comprises a retrieval material information storage device 23, wherein the retrieval material information storage device 23 comprises:

[0102] A term extraction unit 3 extracts terms from the document, i.e., terms contained in a certain page of the document;

[0103] A keyword storage unit 5 stores terms that are keywords associated with terms in the material;

[0104] The keyword extraction unit 7 uses the terms in the document extracted by the term extraction unit 3 and extracts a plurality of keywords associated with the terms in the document from the keyword storage unit 5;

[0105] A subject word storage unit 9 stores subject words associated with keywords;

[0106] The subject word extraction unit 11 uses the multiple keywords extracted by the keyword extraction unit 7 and extracts subject words associated with the keywords from the subject word storage unit 9;

[0107] The search term candidate extraction unit 13 extracts the search term candidates for a certain page of the document from the subject words extracted by the subject word extraction unit 11 and the plurality of keywords extracted by the keyword extraction unit 7;

[0108] The search term candidate display unit 17 displays the search term candidates extracted by the search term candidate extraction unit 13 on the display unit 15;

[0109] a search term input unit 19 for receiving an input indicating a search term among the search term candidates displayed on the display unit 15; and

[0110] The document search information storage unit 21 associates the search term input by the search term input unit 19 with information related to a certain page of the document and stores the information.

[0111] like Fig.12 As shown, the search document information storage device includes a term extraction unit 3, a keyword storage unit 5, a keyword extraction unit 7, a subject word storage unit 9, a subject word extraction unit 11, a search term candidate extraction unit 13, a search term candidate display unit 17, a search term input unit 19, and a document search information storage unit 21. Each means is a means by a computer, and each process is achieved by the cooperation of hardware and software.

[0112] The term extraction unit 3 is a means for extracting terms from data, where the terms in the data are terms contained in a certain page of the data. Examples of data include so-called presentation data. The format of the presentation data is not particularly limited. Examples of presentation software include Microsoft Corporation (registered trademark) PowerPoint (registered trademark), Kingsoft Software Co., Ltd. (registered trademark) WPS OFFICE (registered trademark), Apache (registered trademark) OpenOffice Impress (registered trademark), Keynote (registered trademark), Lotus Freelance (registered trademark), Illustrator (registered trademark), PDF (registered trademark) and Prezi (registered trademark). Examples of data include, for example, data produced using any of these presentation software. Presentation software is software that can display the content of each page in a display unit such as a screen.

[0113] Figure 4 is an example of a page from a presentation. Figure 4As shown, the presentation material includes multiple texts input by the creator. The user can recognize multiple characters with his eyes. On the other hand, the computer stores the text input by the user or the input information related to the text (text size, text color, whether there is animation) and other information together with the text. A preferred example of the term extraction unit 3 is to assign a score (score) to the text according to the input information related to the text (text size, text color, whether there is animation) when extracting the text. For example, the larger the text, the more likely it is to represent the content of the presentation material, so a high score is assigned. In addition, for example, when the text color is red or the text is animated, it represents more content in the presentation material, so a high score is assigned. The term extraction unit 3 pre-stores the score (score) for the effect associated with the text, and reads the score associated with the text when extracting the term. When calculating the score described later, it can be added or multiplied with other scores and scored.

[0114] On the other hand, the term extraction unit 3 itself is a well-known technology. The briefing material has a plurality of text information. Then, the briefing material is stored in a storage unit such as a server (or computer). The term extraction unit 3 reads each page of the stored briefing material and reads the text contained in each page. Then, the term extraction unit 3 performs word class analysis on the read text. At this time, for example, there is a word class database in the storage unit and various terms and their word classes are stored. At this time, the scores of the search terms used as various terms can also be stored in the storage unit according to the purpose. For example, if the data information storage device for retrieval is used for a drug manufacturer, MR (medical information manager), or MS (drug wholesaler), higher scores can be assigned to various disease names compared to general nouns. In addition, although the importance of various drug names or active ingredients is lower than that of disease names, higher scores can also be assigned to general nouns. Then, the term extraction unit 3 can extract the terms (especially nouns) contained in the text, and can use its frequency or the term score stored in the storage unit to extract one or more terms in the data. For example, the term extraction unit 3 extracts term A, term B, and term C from a certain page, term C appears twice and terms A and B appear once, and the scores of terms A, B, and C stored in the storage unit are 5, 50, and 40, respectively, the scores of terms A, B, and C can be set to 5, 50, and 80, respectively. For example, when the number of extracted terms in the document is set to 2, the term extraction unit 3 can extract terms C and B as terms in the document. Then, the extracted terms in the document (terms C and B) are associated with information related to the number of pages that can read the page, and stored in the storage unit. In this way, terms C and B can be read together with the page. Another example of the term extraction unit 3 is to identify a portion of a certain page of the presentation that uses the largest font. Then, a predetermined coefficient is assigned to the terms in the document included in the portion that uses the largest font. The coefficient (first coefficient: a1) can be stored in the storage unit. The term extraction unit 3 stores the first coefficient in the storage unit together with the term in the document included in the portion using the largest font size. Alternatively, the term extraction unit 3 may store a coefficient (second coefficient: a2) according to the font size together with the term in the document.

[0115] The keyword storage unit 5 is a means for storing terms, which are keywords associated with terms in the data. The keyword storage unit 5 can be realized by a storage unit and a requirement (such as a control program) for reading information from the storage unit. Keywords are not only terms in multiple documents, but also terms that are used to easily search each page by using terms related to these as search terms when searching each page. As a result, the number of search terms associated with each page and stored is reduced, which makes it possible to search quickly. However, there are also cases where the terms in the data are directly keywords. Keywords can be called first conversion terms related to the terms in the data. Keywords can also be selected from terms in multiple types of data to be suitable for use in retrieval.

[0116] The terms in the data are the terms included in the brief. Therefore, the terms in the data may not necessarily be consistent with the search terms or may not be suitable as search terms. For example, the brief contains terms called ob gene or ob / ob mouse. They are associated with obesity genes (and obesity, obesity experimental animals). Therefore, the keyword storage unit 5 associates and stores the obesity gene (and obesity, obesity experimental animals) as keywords with the ob gene or ob / ob mouse as the terms in the data.

[0117] Since the keyword storage unit 5 exists, the search terms stored in association with each page become unified terms, so that the number of related pages can be quickly read when searching.

[0118] Figure 5 2 is a conceptual diagram showing a storage example of a keyword storage unit. Figure 5 As shown, the keyword storage unit associates and stores one or more keywords with each term in a plurality of documents, and associates and stores scores (the score is b1) with each keyword. The scores are preferably pre-entered so that the more suitable the term is for search, the higher the score is.

[0119] The keyword extraction unit 7 is a means for using the document terms extracted by the term extraction unit 3 and extracting a plurality of keywords associated with the document terms from the keyword storage unit 5. The keyword storage unit 5 stores terms that are keywords associated with the document terms. Therefore, the keyword extraction unit 7 can use the document terms and read the terms that are keywords associated with the document terms from the keyword storage unit 5. Usually, a plurality of document terms are extracted from a certain page. Therefore, usually, a plurality of terms that are keywords related to this page are extracted. In addition, there are usually a plurality of terms that are keywords associated with the document terms (scores can also be assigned to each of them). Therefore, usually, a plurality of terms that are keywords related to this page are extracted. In addition, of course, there are also cases where the document terms are keywords. That is, the document terms can also be directly extracted as keywords. The keyword extraction unit 7 can use the coefficients of the document terms and the scores of the keywords stored in the storage unit and evaluate the scores of individual keywords. An example of the score of a keyword is a1×a2×b1. In order to calculate the score, the storage unit stores a control program for performing the above operation. Therefore, the control unit can read the control program and read the coefficients and scores stored in the storage unit at the same time, and perform the operation of a1×a2×b1 in the operation unit, and store the operation result in the storage unit. In addition, the frequency of occurrence of the term in the data (the coefficient is a 21 ) or the addition coefficient when extracting a specific keyword from multiple types of data (the coefficient is a 22) is stored in advance in the storage unit, and by calculating a1×a2×a 21 ×a 22 ×b1, thereby calculating the score of the keyword and storing it in the storage unit. In addition, a stronger coefficient can be assigned to the emphasized color contained in a certain page. In this case, there can be a means for analyzing the color of the term from the page, and a storage unit for storing the coefficient of each color, and the color of the term that has been analyzed can be used to read the coefficient related to the color from the storage unit. In addition, not only keywords, but also subject words and classification terms basically store coefficients or scores for various requirements in advance, and read them and perform multiplication or addition operations to calculate the scores, store the scores of individual languages ​​and compare them, thereby calculating the candidates for priority.

[0120] The keyword storage unit 9 is a means for storing keyword-related keywords. The keyword storage unit 9 can be realized by a storage unit and a component (eg, a control program) for reading information from the storage unit.

[0121] For example, in the subject word storage unit, a subject word named obesity can be associated with keywords named obesity gene, obesity, obesity experimental animals, etc. and stored. Subject words can be terms that further unify multiple keywords or superordinate conceptual terms. By using subject words, retrieval can be performed more quickly. Examples of subject words include disease names, drug names, active ingredient names, and pharmaceutical company names. In other words, subject words can be called second conversion terms related to terms in the data. Subject words can also be terms suitable for retrieval for multiple types of keywords. In addition, subject words can also be information related.

[0122] The subject word extraction unit 11 is a means for extracting subject words associated with the keywords from the subject word storage unit 9 using the plurality of keywords extracted by the keyword extraction unit 7 .

[0123] The subject word storage unit 9 stores subject words associated with the keywords. Therefore, the subject word extraction unit 11 uses the plurality of keywords extracted by the keyword extraction unit 7 and extracts subject words associated with the keywords from the subject word storage unit 9.

[0124] Figure 6 2 is a conceptual diagram showing a storage example of a keyword storage unit. Figure 6 As shown, the subject word storage unit associates and stores one or more subject words for each of the plurality of keywords, and associates and stores scores for the individual subject words. The scores are preferably pre-entered so that the more suitable the term is for search, the higher the score is.

[0125] The search term candidate extraction unit 13 is a means for extracting candidates of search terms for a certain page of a document from the subject terms extracted by the subject term extraction unit 11 and the plurality of keywords extracted by the keyword extraction unit 7 .

[0126] For example, a subject word associated with a certain page is stored in one or more storage units. In addition, a plurality of keywords associated with a certain page are stored.

[0127] For example, when the control program controls the subject words as candidates for all search terms and the keywords as candidates for several search terms (for example, considering that the size displayed on the display unit is four), the search term candidate extraction unit 13 uses the read subject words as candidates for all search terms and uses four keywords among the keywords as candidates for search terms.

[0128] In addition, the keyword storage unit 5 may store multiple keywords in association with scores of individual keywords, and the keyword extraction unit 7 may extract scores of individual keywords while extracting multiple keywords. In this case, for example, keywords with higher scores are extracted as candidates for search terms.

[0129] The subject word storage unit 9 may also associate and store subject words with scores of individual subject words, and the subject word extraction unit 11 may also take a predetermined number of keywords (one or more than two) with higher scores among the multiple keywords extracted by the keyword extraction unit 7 as strong subject word candidates, and extract subject words associated with the predetermined number (one or more than two) of strong subject word candidates from the subject word storage unit 9.

[0130] The search material information storage device may further include a classification term storage unit 25 and a classification term extraction unit 27 .

[0131] The classification term storage unit 25 is a means for storing classification terms associated with subject words.

[0132] The classification term extraction unit 27 is a means for using the subject terms extracted by the subject term extraction unit 11 and extracting classification terms associated with the subject terms from the classification term storage unit 25. The classification term can be called a third conversion term related to the term in the data. The classification term can also be a term suitable for classification search selected from multiple types of subject terms. Examples of classification terms can also be objects that show interest in the data. For example, if a page of the data is a certain diabetes drug used by MR (this can be, for example, associated with the subject term and stored), examples of classification terms can also be "MR", "diabetes", "medicine". In addition, if a page of the data is accounting information for bank clerks, examples of classification terms can be bank clerks, etc., or "bank clerks" and "accounting", etc. In addition, the classification term can be information related to the product. Next, the search term candidate display unit 17 of the search data information storage device further extracts the classification term extracted by the classification term extraction unit 27 as one of the candidates for the search term. Figure 7 This is a conceptual diagram showing a storage example of a classification term storage unit. The classification term storage unit associates and stores one or more classification terms with each of a plurality of subject words, and associates and stores scores with individual classification terms. The scores are preferably input in advance so that the more suitable the term is for use in the search, the higher the score is.

[0133] Figure 8 It is a concept map showing the extracted (classification terms), subject terms, keywords, and terms used in the data.

[0134] The search term candidate extraction unit 13 may also extract a predetermined number of keywords (one or more than one) with higher scores from the multiple keywords extracted by the keyword extraction unit 7 as candidates for the search terms. In addition, the search term candidate extraction unit 13 may also use the scores of keywords and the scores of subject terms, and extract a predetermined number (one or more than one) of search term candidates from the subject terms extracted by the subject term extraction unit 11. For example, the subject term storage unit 9 associates and stores the subject terms with the scores of individual subject terms. In addition, the keyword storage unit 5 associates and stores multiple keywords with the scores of individual keywords. A subject term has a keyword from which it is derived. That is, a subject term is read using a keyword. A subject term is often associated with one or more keywords. At this time, the search term candidate extraction unit 13 reads the scores of each keyword from the keyword storage unit 5 while reading the score related to a certain subject term from the subject term storage unit 9. Next, for example, when there are multiple keywords for a certain subject word, the search term candidate extraction unit 13 causes the calculation unit to sum up the scores of the keywords and simultaneously multiply the score of the subject word by the score of the keyword (or the total score of the keywords). By performing the above-described operation, the total score for the subject word is obtained and stored in the storage unit. The search term candidate extraction unit 13 reads the total scores for multiple subject words and causes the calculation unit to compare and extract a predetermined number (one or more than two) of subject words. By performing the above-described operation, the search term candidate extraction unit 13 can extract a predetermined number of subject words even if the number of subject words to be extracted has been determined.

[0135] The search term candidate display unit 17 is a means for displaying the search term candidates extracted by the search term candidate extraction unit 13 on the display unit 15 .

[0136] The search term candidate display unit 17 is:

[0137] While a predetermined number (one or more than one) of keywords extracted as candidates for the search term and a predetermined number (one or more than one) of subject words extracted as candidates for the search term are displayed on the display unit 15 as candidates for the search term,

[0138] The keywords extracted by the keyword extraction unit 7 that are not extracted as search term candidates and the subject words extracted by the subject word extraction unit 11 that are not extracted as search term candidates are displayed on the display unit 15 as preliminary candidates for the search term.

[0139] Search term input unit 19,

[0140] When receiving an input indicating that the search term is to be used as a search term, the preliminary candidates for the search term may be used as the search term.

[0141] The items displayed as search term candidates can be used as search terms, except for those receiving an input indicating that they should not be used as search terms.

[0142] The document search information storage unit 21 is a means for associating and storing the search term input by the search term input unit 19 with the information of a certain page of the document.

[0143] The device of the present invention may further display candidates for content types according to the type of presentation material, and associate and store the type of content with each page of the presentation (or the presentation itself). At this time, the device of the present invention reads the format of the presentation stored in the storage unit (PowerPoint (registered trademark), PDF (registered trademark), Word (registered trademark), etc.). The device of the present invention reads the text contained in the format that has been read. The device of the present invention includes a content analysis term database that has stored terms for content analysis. Next, the device of the present invention uses the terms stored in the content analysis term database to analyze the type of content. For example, when the material is PDF (registered trademark) and a text named "Attached File" exists at the beginning, "Attached File" is extracted as a candidate for the content type of the material. Next, "Attached File" is displayed as a content type on the display unit, and when the user enters confirmation, it is associated with the material, and the "Attached File" related to the content type is stored.

[0144] Fig. 9 is an example of a display screen. In this example, a certain page of the presentation material is displayed in the upper half of the display screen. Next, regarding the candidates for the search terms, icons for acceptance and rejection (check boxes) are displayed, and the candidates for each search term are displayed at the same time. The candidates for the search terms are displayed in Fig. 9 In the example, the data are arranged in the order of classification terms, subject terms, and keywords from the left. The terms used in the data can also be displayed in the display section. Fig. 9 In the example, the term extracted as the search term by the search term candidate extraction unit 13 is checked in the adoption check box. A confirmation button is provided at the bottom of the display screen. When the user uses the confirmation button and inputs a confirmation instruction on the computer (terminal device), the candidate search term is confirmed. The device 1 that receives the input from the computer associates a certain page of the briefing with the confirmed search term (and a score indicating the ease of becoming the search term for each page) and stores it in the storage unit.

[0145] The search term input unit 19 is a means for receiving an input indicating a search term among the search term candidates displayed on the display unit 15 . Fig. 9In the example, the function of the search term input unit 19 is performed by inputting a check box. When the user inputs a candidate of the search term in the adopted state as not adopted, for example, a check is input in the not adopted check box. The device 1 that receives the not adopted input from the check box will change the indicated candidate of the search term to the not adopted state. Then, if the user uses the confirmation button and inputs a confirmation instruction in the computer (terminal device), the candidate of the search term is not adopted. In addition, after the device 1 reduces the score of the candidate of the search term that is not adopted (for example, the score is halved), it is stored as a search term associated with the above-mentioned page. The search term candidate extraction unit 13 checks the not adopted check box (or does not check any check box) for the terms that are not extracted as search terms. When the user inputs a candidate of the search term that is not adopted as adopted, for example, a check is input in the adopted check box. The device 1 receiving the adoption input from the check box will make the indicated search term candidate into the adopted state. Then, if the user uses the confirmation button to input the confirmation instruction on the computer (terminal device), the search term candidate will be adopted. That is, the search term will be associated with this page as the search term of a certain page and stored. At this time, the search term selected by the user can also be stored in a state where the score is added or multiplied.

[0146] Fig.10 This is a flowchart for explaining an example of using the retrieval material information storage device of the present invention. That is, this figure is a figure for explaining a retrieval material information storage method using the retrieval material information storage device. In the figure, S represents a process (step).

[0147] The user creates a presentation material (S101). In this way, the user's terminal device or computer stores the presentation material in a storage unit (or a storage unit of a server).

[0148] The device 1 extracts the terms contained in each page of the briefing material, that is, the terms in the material (S102). At this time, the device 1 can also assign scores to the terms in the material. For example, when the frequency of occurrence of the terms in the material is high, or when accompanied by bold characters, colored text, animations, etc., points can be added in advance and logged in, and the logged-in point information can be used to assign scores to the terms in the material. In addition, the device 1 can have a dictionary of terms in the material, and in the dictionary, for various terms in the material, the terms in the material can be associated with scores and stored, and the device 1 can also read the scores of the terms in the material. Afterwards, the scores of the terms in the material existing in the dictionary and the scores related to the points (such as addition or multiplication) can also be used to calculate the scores of the terms in the material. At this time, when the number of terms in the material is pre-set, the one with the higher score can be used as the term in the material.

[0149] The device 1 uses the terms extracted from one or more documents and extracts a plurality of keywords associated with the terms in the documents from the storage unit (S103). The storage unit records the terms as keywords associated with the terms in the documents. Therefore, the device 1 can use the terms in the documents to extract keywords related thereto from the storage unit. At this time, a score as a search term can be assigned to each keyword. In addition, when the same keyword is selected from terms in different documents, the possibility that the keyword will become a search term is high, so it can also be used as a point-adding object. In this case, points for keywords with higher frequency can be pre-registered, and points based on the number of times the keyword is repeated can be read, and the scores can be added or multiplied. In this way, the scores of multiple keywords (and individual keywords) can be calculated.

[0150] If the device 1 uses a plurality of keywords and extracts subject words associated with the keywords from the storage unit ( S104 ), this process is the same as the keyword extraction step.

[0151] The device 1 can use the extracted subject words and extract classification terms associated with the subject words from the storage unit (S105). This process is an arbitrary process.

[0152] The device 1 extracts candidates for search terms for a certain page of the document from the subject words and multiple keywords (and classification terms) (S106). The device 1 may pre-store control instructions for extracting candidates for search terms, and may extract candidates for search terms for a certain page of the document from the subject words and multiple keywords (and classification terms) according to the control instructions. An example of a control instruction is to extract four keywords with higher scores among multiple keywords and two keywords with higher scores among the subject words (and all classification terms) as candidates for search terms. In this way, candidates for search terms for a certain page of the briefing document are automatically extracted. In addition, the storage unit may also store the extracted candidates for search terms as search terms for a certain page.

[0153] Next, when the user confirms or decides the search term, the device 1 may display the extracted search term candidates on the display unit (S107). At this time, the target page (reduced version) of the presentation, the subject words that are not candidates for the search term, and a plurality of keywords (and classification terms) may also be displayed on the display unit. In this case, the user can select the search term.

[0154] When the user directly confirms, the terminal device receives the confirmation input, and directly stores the search term candidates extracted by the apparatus 1 as search terms associated with a certain page of the presentation material in the storage unit ( S111 ).

[0155] On the other hand, when the terminal device receives input of a candidate that does not use the search term, or when the terminal device receives input of a candidate term that is not a search term, the candidate search terms reflecting these corrections are used as search terms associated with a page in the storage unit (S121).

[0156] After the correction is made, if the user has confirmed it, the terminal device receives the input of the confirmation and stores the corrected search term candidate in the storage unit as a search term associated with a certain page of the presentation material (S122).

[0157] The present invention also provides a search material information storage program for causing a computer to function as the following means or a computer-readable recording medium storing the program.

[0158] The means include:

[0159] A term extraction unit 3 extracts terms from the document, i.e., terms contained in a certain page of the document;

[0160] A keyword storage unit 5 stores terms that are keywords associated with terms in the material;

[0161] The keyword extraction unit 7 uses the terms in the document extracted by the term extraction unit 3 and extracts a plurality of keywords associated with the terms in the document from the keyword storage unit 5;

[0162] A subject word storage unit 9 stores subject words associated with keywords;

[0163] The subject word extraction unit 11 uses the multiple keywords extracted by the keyword extraction unit 7 and extracts subject words associated with the keywords from the subject word storage unit 9;

[0164] The search term candidate extraction unit 13 extracts the search term candidates for a certain page of the document from the subject words extracted by the subject word extraction unit 11 and the plurality of keywords extracted by the keyword extraction unit 7;

[0165] The search term candidate display unit 17 displays the search term candidates extracted by the search term candidate extraction unit 13 on the display unit 15;

[0166] a search term input unit 19 for receiving an input indicating a search term among the search term candidates displayed on the display unit 15; and

[0167] The document search information storage unit 21 associates the search term input by the search term input unit 19 with information related to a certain page of the document and stores the information.

[0168] Fig.11This is a conceptual diagram (block diagram) for illustrating an example of using the retrieval data information storage device of the present invention. In this example, the basic database (DB) includes a content DB, a customer DB, a record DB, and a DB that stores other information. Next, these databases are connected to an engine called the Interactive-PRO framework through an interface. This engine can send and receive information with various terminal devices (such as tablets, mobile terminal devices, and mobile phones) through an application programming interface (API). In addition, this engine can send and receive information with user-side control programs or applications, HTML materials, animation materials, PowerPoint materials, PDF materials, file materials, and database management software. In addition, this engine can synchronize with a server (cloud) and send and receive information. On the other hand, Fig.11 In the example of , this server can send and receive information with various databases and software including customer BI (business intelligence), CRM (customer relationship management), and DWH (data warehouse).

[0169] The present invention also provides a program for causing a computer to function as a display device, or a computer-readable recording medium having the program recorded thereon.

[0170] The display device has the following means, so that the selected words can be displayed on the display unit 69:

[0171] A voice recognition unit 53 for performing voice recognition;

[0172] The conversation-derived word extraction unit 55 extracts a plurality of conversation-derived words from the conversation information recognized by the voice recognition unit 53;

[0173] The search keyword storage unit 57 associates and stores the conversation-derived words with the search keywords;

[0174] The search keyword extraction unit 59 uses the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit 55 and extracts a plurality of search keywords from the search keyword storage unit 57;

[0175] The document storage unit 61 associates and stores each page of the plurality of presentation documents with a search term for each page and a score indicating the ease of becoming the search term for each page;

[0176] The matching page information extraction unit 63 uses the search keyword extracted by the search keyword extraction unit 59 as a search term, and extracts a page of the presentation material associated with the search keyword from the material storage unit 61;

[0177] The selection term extraction unit 65 extracts the search keyword as a selection term for selecting a slide when there is a page matching the presentation material extracted by the page information extraction unit 63; and

[0178] The selected term display unit 71 is for displaying the selected term extracted by the selected term extraction unit 65 on the display unit 69 .

[0179] Industrial Applicability

[0180] The present invention can be used in the display terminal equipment industry, information communication industry, software development, pharmaceutical industry, and financial industry, etc.

[0181] Description of Reference Numerals

[0182] 1. Data storage device for retrieval

[0183] 3. Terminology Extraction

[0184] 5 Keyword storage unit

[0185] 7 Keyword Extraction Department

[0186] 9. Keyword storage unit

[0187] 11. Keyword extraction department

[0188] 13 Search term candidate extraction section

[0189] 15 Display

[0190] 17 Search term candidate display area

[0191] 19 Search term input section

[0192] 21 Data retrieval information storage department

[0193] 23. Data storage device for retrieval

[0194] 25 Classification word storage unit

[0195] 27 Classification term extraction unit

[0196] 53 Voice Recognition Unit

[0197] 55 From conversational phrase extraction

[0198] 57 Search keyword storage unit

[0199] 59 Search keyword extraction unit

[0200] 61 Data Storage Department

[0201] 63 Matching page information extraction unit

[0202] 65 Selected term extraction unit

[0203] 69 Display

[0204] 71 Select term display area

Claims

1. A demonstration device, characterized in that: have: A sound recognition unit (53) for performing sound recognition; A conversation-derived term extraction unit (55) extracts a plurality of conversation-derived terms from the conversation information recognized by the voice recognition unit (53); A search keyword storage unit (57) associates and stores the conversation-derived words with the search keywords; A search keyword extraction unit (59) extracts a plurality of search keywords from the search keyword storage unit (57) using the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit (55); A data storage unit (61) stores each page of a plurality of presentation materials in association with a search term for each page and a score indicating the ease of becoming the search term for each page; and A corresponding page information extraction unit (63) uses the search keyword extracted by the search keyword extraction unit (59) as the search term, and extracts a page of a presentation document associated with the search keyword from the document storage unit (61).

2. The demonstration device according to claim 1, characterized in that: Further having: A selection term extraction unit (65) extracts the search keyword as a selection term for selecting a slide when the presentation material page extracted by the matching page information extraction unit (63) exists; and A selection word display unit (71) for displaying the selection word extracted by the selection word extraction unit (65) on a display unit (69); Thus, the selected word can be displayed on the display unit (69). When the length of the selection word portion displayed on the display unit (69) is shorter than the display portion, one selection word is added.

3. A computer-readable recording medium having a program recorded thereon, characterized in that: The program is for causing the computer to function as a device having the following components: A sound recognition unit (53) for performing sound recognition; A conversation-derived term extraction unit (55) extracts a plurality of conversation-derived terms from the conversation information recognized by the voice recognition unit (53); A search keyword storage unit (57) associates and stores the conversation-derived words with the search keywords; A search keyword extraction unit (59) extracts a plurality of search keywords from the search keyword storage unit (57) using the plurality of conversation-derived terms extracted by the conversation-derived term extraction unit (55); A data storage unit (61) associates and stores each page of a plurality of presentation materials with a search term for each page and a score indicating the ease of becoming the search term for each page; a corresponding page information extraction unit (63) which uses the search keyword extracted by the search keyword extraction unit (59) as the search term and extracts a page of presentation material associated with the search keyword from the material storage unit (61); and A selection term extraction unit (65) extracts the search keyword as a selection term for selecting a slide when the presentation material page extracted by the corresponding page information extraction unit (63) exists.

Images