JP5552987B2 - Search result output device, search result output method, and search result output program - Google Patents

Description

  The present invention relates to a search result output device that outputs a search result, a search result output method, and a search result output program.

  In recent years, there is a growing need for personal computer (PC) use by senior users (hereinafter referred to as senior users). Unlike general users, senior users have problems specific to seniors, such as the low literacy of using PCs and the inability to see small letters.

  For example, consider a task of examining a web page related to “hot spring travel” from the Internet and finally making a travel plan. In general, a search process as shown in FIG. 1 is performed. FIG. 1 is a flowchart of an example of a Web page search process. For example, the idea that a senior user wants to take a hot spring trip triggers the search.

  In step S1, the senior user inputs a search keyword such as “hot spring trip” to the search engine. In step S2, the PC displays a search result display screen as shown in FIG. 2 on a display device or the like.

  FIG. 2 is an image diagram of an example of a search result display screen. The search result display screen of FIG. 2 is an example of a search result display screen of a general search engine. As shown in FIG. 2, the search result display screen displays a list of character-based storage locations of Web pages that have been hit by a search (including search keywords) and an outline of the Web page.

  The senior user selects the Web page by examining the search result display screen of FIG. In step S3, the PC displays the Web page selected by the senior user on a display device or the like. A senior user browses a Web page displayed on a display device or the like, and determines whether the intention matches or is different from the intention.

  If it matches the intention, the senior user achieves the purpose and ends the processing of the flowchart of FIG. If it does not match the intention, the senior user examines the search result display screen again and selects another Web page in step S2. If the search result display screen does not include a web page that matches the intention, the senior user starts over from the input of the search keyword in step S1.

  However, the search result display screen of FIG. 2 is character-based, and there are many information items for senior users. For this reason, it has been difficult for senior users who often have weak visual acuity, judgment, memory, etc. to select Web pages in step S2.

  Conventionally, when browsing web information accumulated by an autopilot or the like, it is possible to easily view web information by generating and outputting thumbnails of web information, outputting only images, and displaying them sequentially. The technique to do is known (for example, refer patent document 1).

JP 2000-112856 A

  However, when browsing Web information stored by an autopilot or the like, a thumbnail of the Web information is generated and output, the contents of the thumbnail are limited to images included in the Web information, and an image of the Web information is displayed. The technique of displaying Web information in a simple manner by sequentially displaying the portions that have been processed cannot solve the above-described problem that it is difficult to perform selection of Web pages.

  For example, even if only a thumbnail of a web page or an image included in the web page is output, a senior user who often has weak visual acuity, judgment, memory, etc., should select the web page in step S2. It was difficult.

  An object of the present embodiment is to provide a search result output device, a search result output method, and a search result output program that can output search results in an easy-to-understand manner.

In order to solve the above-described problem, an embodiment of the present invention relates to an image extracting unit that extracts a representative image representing the content from the content searched by the search keyword, and to the representative image from the searched content. a text extracting unit for extracting the associated text, the extracted representative images, and possess a search result output means for combining the related text associated with the representative image the extracted sequentially output as the search result by the search keyword The search result output means judges the extracted representative image that the user is interested in and the related text related to the extracted representative image based on the user operation when there is a user operation for changing the output speed of the search result. And associating the extracted representative image and related text related to the extracted representative image. The extracted representative image of interest to the user from the recording means to be recorded and the related text related to the extracted representative image are selected as a key image and key text, and the key image and the recording means are recorded. Similarity with the representative image is calculated and output in the order of the similarity .

  In addition, what applied the component, the expression, or the arbitrary combinations of the component of one Embodiment of this invention to a method, an apparatus, a system, a computer program, a recording medium, a data structure, etc. is also effective as an aspect of this invention. .

  According to this embodiment, it is possible to provide a search result output device, a search result output method, and a search result output program that can output search results in an easy-to-understand manner.

It is a flowchart of an example of a web page search process. It is an image figure of an example of a search result display screen. It is an image figure of an example of the search result display screen by a present Example. It is a block diagram of one Example of the web page search system of a present Example. It is a hardware block diagram of an example of a computer. It is a processing block diagram of one Example of a Web server. It is a flowchart of an example of the process in which a Web server displays a search result on a user interface. It is a flowchart of an example of a process of a page analysis part. It is a flowchart of an example showing the image extraction process of step S23. It is a flowchart of an example showing the similar image extraction process of step S26. It is an example flowchart showing the text extraction process of step S24. It is a flowchart of an example showing the similar text extraction process of step S27. It is a flowchart of an example of a process of a synthetic | combination process part. It is a flowchart of an example of a process of a display control part. It is explanatory drawing of an example showing the process image of speech synthesis. It is explanatory drawing of an example showing the processing image of timeline creation and update. It is explanatory drawing of an example showing the process image of step S77-S79. It is explanatory drawing of an example showing the process image of the display control part. It is explanatory drawing of an example showing the processing image of the display control part which received user operation of reproduction speed change. It is explanatory drawing of an example showing the processing image of the display control part which received user operation of reproduction speed change. It is explanatory drawing of an example showing the process image of step S62-S65.

  Next, modes for carrying out the present invention will be described based on the following embodiments with reference to the drawings. In this embodiment, browsing technology in the information search field is taken as an example. In this embodiment, when a search result is output, it is easy to understand what content is written on the Web page indicating the search result. A web page is an example of content (resources) including text and images.

  In the present embodiment, as a search result output method, as shown in FIG. 3, an image representative of each Web page and text related to the image are slide-displayed in a picture-story style. The method of this embodiment is considered suitable for senior users, for example. Furthermore, in this embodiment, not only visual information, but also voice information that is read aloud and BGM that makes Web pages aware of switching are also presented at the same time to promote deeper understanding. it can.

  In order to create a slide representing the contents of a Web page as shown in FIG. 3, a representative image (hereinafter referred to as a representative image) representing the contents of the Web page is extracted from the Web page, and the representative It is necessary to extract a short text (hereinafter referred to as related text) describing the image.

  In general, a Web page includes various images such as an image such as an advertisement banner, an icon image for character decoration, and an image such as a photograph for explaining text in the Web page. Yes.

  When the user selects a desired Web page from the display of the search result, the Web pages are compared, and those closer to the desired Web page are narrowed down. For example, when a representative image and related text are extracted from a Web page, the representative image and related text can be selected based on a uniform standard. However, since the representative image and the related text to be noticed are actually different depending on the user, it is more useful to select the Web page if the representative image and the related text in which the user is interested are displayed. That is, it is necessary to associate and extract the representative image and the related text that the user has shown interest in.

  FIG. 4 is a configuration diagram of an embodiment of the Web page search system of the present embodiment. 4 includes one or more clients 1a to 1n, one or more Web servers 2a to 2m, and a network 3. In addition, when any of the clients 1a to 1n may be used, the clients 1 are collectively referred to. When any of the Web servers 2a to 2m may be used, the Web server 2 is collectively referred to.

  The client 1 is connected to the Web server 2 via the network 3 so that data communication is possible. The client 1 is a terminal operated by a user. The web server 2 stores web pages. Further, at least one of the Web servers 2 includes a search engine.

  When there is a search request specifying a search keyword from the client 1, the Web server 2 including the search engine searches a public Web page with the search keyword, and sends the search result Web page to the client 1 via the network 3. Send.

  Further, when the Web server 2 receives a Web page acquisition request specifying the Web page storage location from the client 1, the Web server 2 transmits the Web page for which the acquisition request has been made to the client 1 via the network 3. The storage position of the Web page is represented by a URL (Uniform Resource Locator) or the like.

  For example, the client 1 designates a search keyword and makes a Web page search request to the Web server 2. Further, the client 1 specifies a URL, for example, and requests a Web page from the Web server 2. The client 1 displays a web page received from the web server 2 using a web browser or the like.

  The client 1 and the Web server 2 are realized by a computer 10 such as a server computer or a client computer. The computer 10 in which the search result output program is installed is an example of a search result output device.

  FIG. 5 is a hardware configuration diagram of an example of a computer. Here, the hardware configuration of the computer 10 in which the search result output program is installed will be described. The computer 10 executes a search result output program having the same function as the search result output device.

  The computer 10 in FIG. 5 includes an input device 21, a display device 22, and a computer main body 23. The computer main body 23 includes a main storage device 31, an arithmetic processing device 32, an interface device 33, a recording medium reading device 34, and an auxiliary storage device 35 that are connected to each other via a bus 37. The input device 21 and the display device 22 are connected to the bus 37.

  The input device 21, the display device 22, the main storage device 31, the arithmetic processing device 32, the interface device 33, the recording medium reading device 34, and the auxiliary storage device 35 that are mutually connected by the bus 37 are managed by the arithmetic processing device 32. Can exchange data with each other. The arithmetic processing unit 32 is a central processing unit that controls operation of the entire computer 10.

  The interface device 33 receives data from the network 3 or the like and passes the contents of the data to the arithmetic processing device 32. The interface device 33 transmits data to the network 3 or the like in response to an instruction from the arithmetic processing device 32.

  The auxiliary storage device 35 stores a search result output program that causes the computer 10 to execute at least processing in the search result output device as part of a program that causes the computer 10 to perform the same function as the search result output device. Then, when the arithmetic processing device 32 reads out the search result output program from the auxiliary storage device 35 and executes it, the computer 10 functions as a search result output device. The search result output program may be stored in the main storage device 31 accessible to the arithmetic processing device 32.

  The input device 21 receives data input under the control of the arithmetic processing device 32. The search result output program can be recorded in a recording medium 36 that can be read by the computer 10.

  Examples of the recording medium 36 include a magnetic recording medium, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Magnetic recording media include HDDs, flexible disks (FD), magnetic tapes (MT) and the like. Examples of the optical disc include a DVD (Digital Versatile Disc), a DVD-RAM, a CD-ROM (Compact Disc-Read Only Memory), and a CD-R (Recordable) / RW (ReWriteable). Magneto-optical recording media include MO (Magneto-Optical disk).

  When distributing the search result output program, for example, it is conceivable to sell a portable recording medium 36 such as a DVD or CD-ROM in which the search result output program is recorded. For example, the computer 10 that executes the search result output program reads the search result output program from the recording medium 36 on which the recording medium reader 34 has recorded the search result output program. The arithmetic processing unit 32 stores the read search result output program in the main storage device 31 or the auxiliary storage device 35. Then, the computer 10 reads the search result output program from the main storage device 31 or the auxiliary storage device 35 which is its own storage device, and executes processing according to the search result output program. That is, the arithmetic processing unit 32 implements various processes as described later according to the search result output program.

  Here, the Web server 2 as an example of the search result output device will be described. FIG. 6 is a processing block diagram of an embodiment of the Web server. Note that the processing block diagram of FIG. 6 is an example, and a configuration in which a part of the processing blocks is provided to another Web server 2 or client 1 is also possible.

  The Web server 2 includes a search unit 41, a page analysis unit 42, a composition processing unit 43, and a display control unit 44. Note that the user interface 60 in FIG. 6 is realized by the client 1 and a Web browser, for example.

  The search unit 41 is, for example, a search engine. The search unit 41 receives a search request specifying a search keyword from the user interface 60. The search keyword may be registered in advance, or may be freely described by the user. The search unit 41 searches a public Web page with a search keyword, and obtains a search result (for example, a URL list of the Web page).

  The page analysis unit 42 includes a page acquisition unit 51, an image extraction unit 52, a text extraction unit 53, and a recording unit 54. The page acquisition unit 51 of the page analysis unit 42 acquires a URL list of Web pages as a search result from the search unit 41. The image extraction unit 52 downloads the Web page of the URL list acquired by the page acquisition unit 51.

  As will be described later, the image extraction unit 52 extracts, as a representative image, an image that matches an extraction condition from images in the downloaded Web page (images included in the Web page). The text extraction unit 53 analyzes the HTML of the downloaded web page, and extracts text that matches the extraction condition as related text from text in the web page (text included in the web page), as will be described later. The recording unit 54 records the URL, title, representative image (URL or file) and related text of the Web page included in the URL list in the database.

  The synthesis processing unit 43 performs speech synthesis processing on the related text recorded in the database of the recording unit 54, and generates a read-out voice file. In addition, the synthesis processing unit 43 generates music (for example, BGM) synchronized with the voice reading time for each Web page. The synthesizing processing unit 43 sequentially displays the title, representative image, and related text of the Web page recorded in the database of the recording unit 54, and at the same time a synchronized file (time) for reproducing the read-out voice file and music in synchronization. Line).

  The display control unit 44 slides and reproduces each Web page according to the synchronization file. The display control unit 44 performs control such as stop / play / return / advance of the slide in accordance with a user operation, and changes the slide display order and display contents in accordance with the control as described later. Further, when a web page browsing operation is performed, control is performed to display the web page on the web browser.

  As described above, the Web server 2 in FIG. 6 displays search results as shown in FIG. 7 using the Web browser of the client 1 as the user interface 60. FIG. 7 is a flowchart of an example of processing in which the Web server displays search results on the user interface.

  In step S <b> 11, the page analysis unit 42 performs a later-described page analysis process on the Web page of the URL list acquired as a search result from the search unit 41. In step S12, the composition processing unit 43 creates a reading voice file and a timeline to display the title, representative image, and related text of the Web page recorded in the database of the recording unit 54 in order, A synthesizing process to be described later for synchronously playing music is performed.

  In step S13, the display control unit 44 performs slide playback of each Web page according to the synchronization file, and performs control such as slide stop / playback / return / advance according to the user operation. A display control process to be described later for changing the display contents is performed. In the display control process in step S13, as described later, after the image and text in which the user is interested are set as the key image and key text, the process returns to the page analysis process shown in step S11.

  FIG. 8 is a flowchart of an example of processing of the page analysis unit. In step S21, the page acquisition unit 51 determines whether the key image and key text are set. For example, the key image and the key text are not set when the first activation or no user operation is performed.

  If it is determined that the key image and key text are not set, the page acquisition unit 51 acquires the URL list (page list) of the Web page as the search result from the search unit 41 in step S22.

  In step S <b> 23, the image extraction unit 52 downloads the Web page of the page list acquired by the page acquisition unit 51. As will be described later, the image extraction unit 52 extracts a representative image that matches the extraction condition from the images in the downloaded Web page.

  In step S24, the text extraction unit 53 analyzes the HTML of the downloaded web page, and extracts text that matches the extraction condition from the text in the web page as related text, as will be described later.

  On the other hand, when it is determined in step S21 that the key image and the key text are set, the page acquisition unit 51 notifies the image extraction unit 52 and the text extraction unit 53 that the key image and the key text are set.

  When notified that the key image and key text are set, the image extracting unit 52 acquires all the recorded representative images from the recording unit 54 in step S25. When notified that the key image and key text are set, the text extraction unit 53 acquires all the related texts recorded from the recording unit 54.

  In step S26, the image extraction unit 52 performs similar image extraction processing on all the acquired representative images using a key image as described later. In step S27, the text extraction unit 53 performs similar text extraction processing on all the acquired related texts as described later using the key text.

  FIG. 9 is a flowchart illustrating an example of the image extraction process in step S23. In step S31, the image extraction unit 52 acquires all images included in the downloaded web page.

  In step S32, the image extraction unit 52 checks the types of all acquired images (animation gif, flash, jpeg, etc.) and link attributes (HTML a tag, etc.) and excludes external links and non-target image types. Perform image attribute filtering. In step S33, the image extraction unit 52 checks the image sizes of all the images that have not been excluded by the image attribute filtering process, and performs an image size filtering process that excludes images having an image size less than a certain threshold. In step S33, for example, icons and logo images are excluded by excluding images having an image size of less than 60 × 60 pixels.

  In step S34, the image extraction unit 52 checks the aspect ratio (aspect ratio) of all images that have not been excluded by the image attribute filter processing and the image size filter processing, and excludes images having an aspect ratio equal to or greater than a certain threshold. Perform filtering. In step S34, for example, advertisement banner images are excluded by excluding images with an aspect ratio of 1: 2 or more.

  In step S35, the image extraction unit 52 checks the presence / absence of characters in all the images that have not been excluded by the image attribute filter processing, the image size filter processing, and the aspect ratio filter processing, and excludes images that include characters. Perform detection filter processing. In step S35, images such as campaigns and special features are excluded by excluding images containing characters. In the character detection filter process, the size of the character to be detected may be set to be, for example, less than 20 × 20 pixels.

  In step S <b> 36, the image extraction unit 52 records all images not excluded by the image attribute filter processing, image size filter processing, aspect ratio filter processing, and character detection filter processing in the recording unit 54 as representative images. The image extraction unit 52 sets the recording order in the Web page unit as the Web page rank (page list) order of the search results, and sets the recording order in the representative image unit in each Web page as the image size order.

  FIG. 10 is a flowchart showing an example of the similar image extraction process in step S26. In step S <b> 41, the image extraction unit 52 acquires all representative images recorded in the recording unit 54. In step S <b> 42, the image extraction unit 52 calculates the similarity with the key image for all the acquired representative images. It should be noted that the method for calculating the similarity to the key image can use a commonly used Euclidean distance of a color histogram.

  Also, as an example of the method of calculating the similarity with the key image, for example, “Kenji Suzuki, et.al.,“ Information Database Technology (Future Netto Technology Series) ”, Chapter 3 Multimedia Information Storage / Search Technology, pp .80-136, Ohmsha, 2000 "can be used.

  In step S43, the image extraction unit 52 rerecords all the representative images in the recording unit 54 in the order of similarity with the key image calculated in step S42.

  FIG. 11 is a flowchart illustrating an example of the text extraction process in step S24. The text extraction process in step S24 is, for example, “Endo, Sashida, Masumoto, Nagata, Tanahashi: Interactive Web search system that uses image information and text information in an integrated manner, Intelligent Information Media Symposium, 5th Intelligent Information Media Symposium paper. Pp.163-170 (1999) ”can be used.

  In the above existing method, when acquiring the related text related to the representative image, it is difficult to analyze and acquire the semantic content of the text. The rule of thumb is that they are often “related” and “the distance between an image and text can be estimated to some extent from the tag structure of HTML”.

  In terms of the closeness of the image and the text, it can be roughly obtained by analyzing the contents of the HTML file and acquiring the closeness between the tags in the HTML file. Although there are some exceptions such as tags having an arrangement structure such as a table, generally tags close to each other in an HTML file appear close to each other on the display. Therefore, the measure of the positional proximity between the image and the text uses the number of line breaks sandwiched between them. Thus, the greater the number of lines between the image and the text, the greater the distance.

  Further, the distance between the image and the text by the HTML tag structure is calculated from the type and number of tags included between the image and the text, using the relationship between the tag and the distance value shown in Table 1. If there are many tags in between, the image and text are considered to be far apart. In particular, when there are <P>, <H1>, <HR>, and the like, which are tags representing sentence breaks, it is considered that the distance between the image and the text is greatly separated.

  In addition, as a special case of the distance between the image and the text, the text described in the ALT attribute representing the text substituted for the image attached to <IMG> which is a tag for displaying the image is related to the image. Since the nature is deep, a distance value of 0 is assigned.

  In step S51, the text extraction unit 53 analyzes the HTML of the downloaded web page, and acquires the position of the representative image (the position of the tag for displaying the representative image) in the order of recording in the HTML. In step S52, the text extraction unit 53 expands the acquired text range in tag units from the position of each representative image acquired in step S51.

  When text is included in the acquired text range, in step S53, the text extraction unit 53 calculates the distance value between the text included in the acquired text range and the representative image using the relationship between the tag and the distance value shown in Table 1. To do. In step S54, the text extraction unit 53 determines whether the calculated distance value between the text and the representative image is equal to or less than a threshold value.

  If the calculated distance value between the text and the representative image is equal to or smaller than the threshold value, the text extracting unit 53 records the text whose distance value to the representative image is equal to or smaller than the threshold value in the recording unit 54 in step S55. . If the calculated distance value between the text and the representative image is not less than the threshold value, the text extraction unit 53 further expands the acquired text range in step S52.

  FIG. 12 is a flowchart illustrating an example of the similar text extraction process in step S27. In step S61, the text extraction unit 53 analyzes the HTML of the downloaded Web page, and acquires the position of the representative image in the order of recording in the HTML.

  In step S62, the text extraction unit 53 expands the acquired text range in character units from the position of each representative image acquired in step S61. In step S63, the text extraction unit 53 calculates the similarity between the text included in the acquired text range and the key text. As a method for calculating the similarity with the key text, a commonly used existing method for calculating the similarity of text can be used.

  In step S64, the text extraction unit 53 determines whether the similarity between the text included in the calculated acquired text range and the key text is greater than or equal to a threshold value and is a maximum value.

  If the similarity between the text included in the calculated acquired text range and the key text is equal to or greater than the threshold value and the maximum value, the text extraction unit 53 determines that the similarity with the key text is equal to or greater than the threshold value in step S65. The text having the maximum value is re-recorded (rewritten) in the recording unit 54 as the related text.

  If the similarity between the text included in the calculated acquired text range and the key text is not equal to or greater than the threshold value or is not a maximum value, the text extracting unit 53 further expands the acquired text range in step S62. The processes in steps S62 to S65 in FIG. 12 will be described later.

  FIG. 13 is a flowchart of an example of processing of the synthesis processing unit. In step S71, the composition processing unit 43 acquires the related text recorded in the recording unit 54. In step S72, the synthesis processing unit 43 inputs the acquired related text to the speech synthesis engine, performs speech synthesis processing, and generates a read-out speech file.

  In step S73, the composition processing unit 43 determines whether to use the thumbnail image of the Web page. If the thumbnail image of the Web page is not used, the composition processing unit 43 determines in step S74 whether or not it is the first composition process. Whether or not the synthesis process is the first time can be determined by, for example, the presence or absence of a timeline.

  If it is the first synthesizing process, the synthesizing unit 43 generates music (for example, BGM) synchronized with the voice reading time for each Web page in step S75. Then, the composition processing unit 43 sequentially displays the Web page title, the representative image, and the related text recorded in the recording unit 54, and creates a timeline for reproducing the read-out voice file and the music in synchronization.

  If it is not the first synthesizing process, the synthesizing unit 43 generates music synchronized with the voice reading time for each Web page in step S76. Then, the composition processing unit 43 updates the timeline.

  If it is determined in step S73 that the thumbnail image of the Web page is to be used, the composition processing unit 43 creates a thumbnail image of the Web page that has been rendered in accordance with the user's Web browser in step S77.

  In step S78, the composition processing unit 43 acquires the representative image on the thumbnail image and the position (coordinates) of the related text. In step S79, the composition processing unit 43 calculates the enlargement ratio and the scroll amount from the display screen size. In step S74, the composition processing unit 43 determines again whether or not it is the first composition processing. The display control unit 44 can display the representative image on the thumbnail image of the Web page and the related taste in order based on the enlargement ratio and the scroll amount calculated in step S79. The process of FIG. 13 will be described later.

  FIG. 14 is a flowchart of an example of processing of the display control unit. In step S80, the display control unit 44 performs slide reproduction along the timeline. In the slide playback performed in step S80, the web page (slide) including the title of the web page, the representative image and the related text recorded in the recording unit 54 is displayed in order along the timeline, and the read-out voice file and music are displayed. Synchronized playback.

  In step S81, the display control unit 44 determines whether or not there has been a user operation for changing the reproduction speed (slide stop / speed down / return, etc.). If there is no user operation for changing the reproduction speed, the display control unit 44 determines in step S82 whether or not there is a browsing user operation.

  If there is no browsing user operation, the display control unit 44 performs the page analysis process again in step S11. If there is a browsing user operation, the display control unit 44 performs browser display processing for displaying a Web page on the Web browser in step S83. It is assumed that the Web page displayed in step S83 is a Web page corresponding to the slide displayed on the Web browser when a browsing user operation is performed.

  If it is determined in step S81 that there is a user operation for changing the playback speed, the display control unit 44 determines in step S84 whether the user operation is “return”. When determining that the user operation is “return”, the display control unit 44 sets the representative image and the related text on the immediately preceding slide as the key image and the key text in step S85.

  If it is determined in step S84 that the user operation is not “return”, the display control unit 44 determines whether the user operation is “speed down” or “stop” in step S86.

  If the display control unit 44 determines that the user operation is “speed down” or “stop”, the representative image and the related text on the current slide are set as the key image and the key text in step S87. The process in FIG. 14 will be described later.

  FIG. 15 is an explanatory diagram showing an example of a speech synthesis process image. The synthesis processing unit 43 acquires the related text related to the representative screen recorded in the recording unit 54, inputs the acquired related text to the speech synthesis engine 100, performs speech synthesis processing, and reads out the voice file (in FIG. 15). Audio files (1) to (3)) are generated.

  FIG. 16 is an explanatory diagram illustrating an example of a process image for creating and updating a timeline. The composition processing unit 43 creates or updates a timeline as shown in FIG. For example, the timeline shown in FIG. 16 includes a representative screen (1) and related text (1), a representative screen (2) and related text (2), a representative screen (3) and related text (3) on a certain Web page. It shows that it displays in order.

  Further, the timeline shown in FIG. 16 reproduces the audio file (1) and BGM (1) as an example of music in synchronization with the representative screen (1) and the related text (1) on a certain Web page. Synchronizing with the representative screen (3) and the related text (2), reproducing the audio file (2) and BGM (1) as an example of music in synchronization with the representative screen (2) and the related text (2) Thus, the audio file (3) and BGM (1) as an example of music are reproduced. Note that BGM (1) is used in the timeline shown in FIG. 16 in order to audibly express the switching of Web pages.

  FIG. 17 is an explanatory diagram illustrating an example of the processing image of steps S77 to S79. The composition processing unit 43 creates a thumbnail image 200 of the Web page that has been rendered in accordance with the user's Web browser. The composition processing unit 43 acquires the positions of the representative images 201 to 203 and the related texts (1) to (3) on the thumbnail image 200. The composition processing unit 43 calculates the enlargement ratio and the scroll amount from the display screen size.

  Based on the calculated enlargement ratio and scroll amount, the display control unit 44 displays the representative image 201 and the related text (1) portion, the representative image 202 and the related text (2) portion, and the representative image 203 on the thumbnail image 200 of the Web page. The related text (3) can be zoomed and displayed in order.

  In other words, the processes of steps S77 to S79 can quickly confirm only the important part for selecting the Web page without destroying the layout of the original Web page.

  FIG. 18 is an explanatory diagram illustrating an example of a processing image of the display control unit. As shown in FIG. 18 along the timeline, the display control unit 44 sequentially displays slides that are Web pages including representative images and related text along the timeline, and synchronizes the read-out voice file and music. Reproduce.

  19 and 20 are explanatory diagrams illustrating an example of a processing image of the display control unit that receives a user operation for changing the reproduction speed. Assume that the display dwell time of a slide represents the level of user interest in that slide.

  FIG. 19 shows an example in which a user operation for changing the playback speed is received using the back / forward / stop button. In the slides 301 and 302, the user presses the “forward” button to change the playback speed. The slide 303 has the playback speed changed by the user pressing the “stop” button.

  The slide 303 is considered to be of great interest to the user because the user has pressed the “stop” button. It is desirable that the slide to be displayed next to the slide 303 has contents related to the representative screen of the currently displayed slide 303 and related text. Therefore, in this embodiment, the representative image of the slide 303 and the related text are set as the key image and the key text as a reference representing the user's interest.

  The slides after the slide 303 are rearranged and displayed in the order of the slides 304 to 306 similar to the slide 303 that is considered to be of high user interest. For example, in FIG. 20, after the slide 303, the slide 304 having the first similarity with the slide 303 is displayed, then the slide 305 with the second similarity with the slide 303 is displayed, and then similar to the slide 303. A slide 306 with the third degree is displayed. Note that a slide with high user interest may be directly selected by the user.

  FIG. 21 is an explanatory diagram illustrating an example of processing images of steps S62 to S65. FIG. 21A shows the ranking order of Web pages included in the search result of the search keyword “hot spring”. FIG. 21B shows the display order of slides expressing the contents of the Web page of FIG. FIG. 21B shows that the slides 404, 405, 406, and 407 are displayed in this order if a user operation for changing the playback speed is not accepted.

  Here, it is assumed that when the slide 405 is displayed, the user changes the playback speed by pressing the “stop” button. In the present embodiment, it is considered that the images similar to each other in the Web pages included in the search result of the same search keyword “hot spring” and the adjacent texts are related to each other (similar contents). Therefore, the representative image P and the related text T of the Web page 402 corresponding to the slide 405 are set as a key image and key text.

  Thereafter, in this embodiment, the representative image of the Web page 403 similar to the representative image P that is a key image is extracted as a similar image of the representative image P. Next, in this embodiment, an extraction range to be extracted as related text is determined using the position of the representative image of the Web page 403 as a starting point. The extraction range evaluates the similarity with the key text while enlarging the preceding and succeeding text in character units with the representative image of the Web page 403 as the center. The range in which the similarity exceeds the threshold and the similarity reaches a maximum value is set as the related text.

  The display order of the slides after the slide 405 whose playback speed has been changed by the user pressing the “stop” button is the order of similarity with the key image. For example, in FIG. 21B, a slide 407 of a representative image (similar image of the representative image P) of the slide 405 is displayed instead of the slide 406 that is scheduled to be displayed next to the slide 405.

  In the present embodiment, the contents of the Web page included in the search result are presented to the user as a representative image, related text, and voice, so that the senior user can easily select the Web page. In particular, according to the present embodiment, the representative image and the related text in consideration of the content of the Web page included in the search result are displayed according to the user's interest, so that the search condition is not newly set again. It becomes easy to narrow down to a Web page desired by the user.

  When the client 1 is a PC, for example, by registering a search keyword of interest in advance, it is possible to easily search for a web page as if watching a TV, for example. The slide reproduction may be performed using, for example, a screen saver. When the client 1 is a small screen with a limited amount of information that can be displayed, such as a mobile phone or a portable information terminal, it is possible to provide a material for determining whether or not to browse a Web page using only a representative image and related text.

The present invention may have the following configurations as described below.
(Appendix 1)
Image extracting means for extracting a representative image representing the content from the content searched by the search keyword;
Text extraction means for extracting related text related to the representative image from the searched content;
A search result output device comprising: a search result output unit that outputs the extracted representative image and related text related to the extracted representative image together as a search result based on the search keyword.
(Appendix 2)
The search result output means outputs the extracted representative image and the extracted representative image that the user is interested in from the recording means that records the extracted representative image and related text related to the extracted representative image in association with each other. A related related text is selected as a key image and a key text, a similarity between the key image and the representative image recorded in the recording unit is calculated, and output in the order of the similarity. The search result output device according to appendix 1.
(Appendix 3)
A search result output method executed by a computer,
A representative image representing the content is extracted from the content searched by the search keyword,
Extracting related text related to the representative image from the retrieved content;
A search result output method comprising: outputting the extracted representative image and related text related to the extracted representative image together as a search result based on the search keyword.
(Appendix 4)
When the extracted representative image and the related text related to the extracted representative image are output together as a search result based on the search keyword, the extracted representative image and the related text related to the extracted representative image Selecting the extracted representative image that is of interest to the user from the recording means that records the information and the related text related to the extracted representative image as a key image and key text,
The search result output method according to appendix 3, wherein the similarity between the key image and the representative image recorded in the recording means is calculated and output in the order of the similarity.
(Appendix 5)
On the computer,
A representative image representing the content is extracted from the content searched by the search keyword,
Extracting related text related to the representative image from the retrieved content;
The search result output program which performs the process which combines the extracted representative image and the related text relevant to the extracted representative image, and outputs it as a search result by the search keyword.
(Appendix 6)
When the extracted representative image and the related text related to the extracted representative image are output together as a search result based on the search keyword, the extracted representative image and the related text related to the extracted representative image Selecting the extracted representative image that is of interest to the user from the recording means that records the information and the related text related to the extracted representative image as a key image and key text,
The search result output program according to appendix 5, wherein a similarity between the key image and the representative image recorded in the recording unit is calculated, and a process of outputting in order of the similarity is executed.

  Note that the content processing program in this embodiment can be provided not only by package software but also by a WEB service or the like.

  The image extraction means described in the claims corresponds to the image extraction section 52, the text extraction means corresponds to the text extraction section 53, and the search result output means corresponds to the synthesis processing section 43 and the display control section 44. The recording unit corresponds to the recording unit 54.

DESCRIPTION OF SYMBOLS 1, 1a-1n Client 2, 2a-2m Web server 3 Network 10 Computer 21 Input device 22 Display device 23 Computer main body 31 Main storage device 32 Arithmetic processing device 33 Interface device 34 Recording medium reader 35 Auxiliary storage device 36 Recording medium 37 Bus 41 Search unit 42 Page analysis unit 43 Composition processing unit 44 Display control unit 51 Page acquisition unit 52 Image extraction unit 53 Text extraction unit 54 Recording unit 60 User interface

Claims (3)

Image extracting means for extracting a representative image representing the content from the content searched by the search keyword;
Text extraction means for extracting related text related to the representative image from the searched content;
The extracted representative images, and possess a search result output means for combining the related text associated with the representative image the extracted sequentially output as the search result by the search keyword,
When there is a user operation for changing an output speed of the search result, the search result output means determines the extracted representative image that the user is interested in and the related text related to the extracted representative image based on the user operation. The extracted representative image and the related text related to the extracted representative image that the user is interested in from the recording unit that records the extracted representative image and the related text related to the extracted representative image in association with each other. A search result selected as an image and key text, calculating a similarity between the key image and the representative image recorded in the recording means, and outputting the similarity in the order of the similarity Output device. A search result output method executed by a computer,
A representative image representing the content is extracted from the content searched by the search keyword,
Extracting related text related to the representative image from the retrieved content;
The extracted representative image and the related text related to the extracted representative image are output in order as a search result by the search keyword ,
When there is a user operation for changing the output speed of the search result, the extracted representative image that the user is interested in based on the user operation and the related text related to the extracted representative image are determined, and the extracted representative image and The extracted representative image of interest to the user and the related text related to the extracted representative image are selected as a key image and key text from a recording unit that records the related text related to the extracted representative image in association with each other. A search result output method comprising calculating a similarity between the key image and the representative image recorded in the recording means and outputting the similarity in the order of the similarity . On the computer,
A representative image representing the content is extracted from the content searched by the search keyword,
Extracting related text related to the representative image from the retrieved content;
The extracted representative image and the related text related to the extracted representative image are output in order as a search result by the search keyword ,
When there is a user operation for changing the output speed of the search result, the extracted representative image that the user is interested in based on the user operation and the related text related to the extracted representative image are determined, and the extracted representative image and The extracted representative image of interest to the user and the related text related to the extracted representative image are selected as a key image and key text from a recording unit that records the related text related to the extracted representative image in association with each other. A search result output program for executing a process of calculating a similarity between the key image and the representative image recorded in the recording means and outputting the similarity in the order of the similarity .