JP2008052579A - Retrieval device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To present a retrieval result so that a user understands important contents while minimizing the user's workload. <P>SOLUTION: A determining section 14 determines and stores the importance (ranks) of each document and pages included in the document based on a display time stored in a display time management section 15, the number of marking stored in a marking number management section 16, and display frequency stored in a display frequency management section 17. The ranks are determined by page of each document. The ranks for every page are further totaled, and the ranks for every document are obtained by a value (an average value) obtained by dividing the total value by the number of ranked pages. In retrieving the document by a keyword, for instance, the ranked result is reflected in the retrieval result. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technique for searching content such as a document.

As the capacity of storage devices has been increased, various contents including documents and videos can be stored in large quantities. In a general-purpose computer, when searching for a desired content from a vast number of contents, the user specifies, for example, the data file name, data format, data size, creation / access date or keyword, and the specified contents In general, a computer searches for content that matches the above. In a search system such as a library, metadata relating to each book is manually input in advance, and the book is searched by specifying the content of the metadata at the time of the search. Further, Patent Document 1 proposes a technique for adding markings such as underline and sirioli to content that the user considers important.
JP-A-9-101971

  In order to obtain a more useful search result in the conventional search method, for example, prior work such as input of metadata is required, which is very complicated and troublesome. In addition, when displaying the search results, the searched contents are simply displayed in a predetermined order. Therefore, when a large number of contents are included in the search results, the user must You must look at the contents and check the importance of each one.

  The present invention has been made in view of the above-described background, and an object of the present invention is to present a search result that can identify content important to a user while reducing the work load of the user as much as possible.

  In order to achieve the above object, the present invention measures the degree to which an image displayed on a display means is browsed based on content data, and generates a browsing degree information representing the degree of browsing, Based on the browsing degree information generated by the measuring means, the determination means for determining the importance of the content data, and the content data that matches the specified search condition are searched from the plurality of content data, and the search There is provided a search device comprising: search means for outputting a result in an output mode corresponding to the degree of importance determined for each content data by the determination means. According to the present invention, when searching for content, it is possible to output a search result that can identify content important to the user. In addition, since the importance of each content data is determined by the act of the user browsing the content, no special work or operation is imposed on the user.

  In a preferred aspect of the present invention, the measuring means is the time when the image is displayed on the display means, the frequency at which the image is displayed on the display means, or the number of information given by the user to the image. At least one of them is measured, and the measured value is used as the browsing degree information.

  In a preferred aspect of the present invention, the measurement means measures the frequency at which the image is displayed on the display means, and uses the measurement value as the browsing degree information, and the determination means determines the frequency as the frequency increases. For the function with higher importance, the browsing degree information generated by the measuring means is input to determine the importance of the content data, and the function uses the importance as the vertical axis and the frequency as the horizontal axis In this case, the function becomes a convex function until the frequency reaches a predetermined value, and the function functions as a convex function when the frequency exceeds the predetermined value. In this way, it is possible to determine the importance according to the display frequency characteristics.

  In a preferred aspect of the present invention, the measuring unit generates a plurality of types of browsing degree information, and the determining unit is configured to determine a browsing level respectively represented by the plurality of types of browsing level information generated by the measuring unit. The degree of importance of the content data is determined by proportional distribution. Considering multiple types of browsing degree information, the determination accuracy of importance can be improved.

  In a preferred aspect of the present invention, the measuring unit generates a plurality of types of browsing degree information, and the determining unit is configured to respond to a length of a target period for determining the importance degree among the plurality of types of browsing level information. The degree of importance of the content data is determined based on different types of browsing degree information. Considering multiple types of browsing degree information, it is possible to improve the determination accuracy of the importance level, and it is possible to determine the importance level according to the length of the target period for determining the importance level.

  In a preferred aspect of the present invention, the content data is document data composed of a plurality of page data, and the measuring means measures the degree of browsing of the page displayed on the display means, and the degree of browsing The degree-of-view information that represents the degree of importance of each page included in each of the documents and the degree of importance of each of the document units based on the degree-of-view information generated by the measuring means The search means searches for document data or page data that matches a specified search condition from a plurality of document data stored in the storage means, and the search result is obtained by the determination means. The document data and each page data are output in an output mode corresponding to the determined importance.

In a preferred aspect of the present invention, the apparatus includes a user specifying unit that specifies a user who browses an image displayed on the display unit, and the measuring unit is a browsing level indicating a level of viewing of the image displayed by the display unit. Storing the information in association with the user identification information assigned to the user specified by the user specifying means as the user viewing the image,
The determination means determines the importance of the content data based on browsing degree information stored in association with each user identification information and stored by the measurement means. In this way, even when a plurality of users browse the displayed image, it is possible to determine the importance according to each user.

  Further, the present invention provides a computer having a determination function for determining the importance of the content data based on browsing degree information indicating a degree to which an image displayed on the display unit is browsed based on the content data, A search function for searching content data that matches a specified search condition from the content data, and outputting the search result in an output mode according to the importance determined for each content data by the determination function; Provide a program to make it happen. According to the present invention, when searching for content, it is possible to output a search result that can identify content important to the user. In addition, since the importance of each content data is determined by the act of the user browsing the content, no special work or operation is imposed on the user.

  In this embodiment, a “document” composed of one or a plurality of pages is set as “content” to be searched. Then, how important each content is by each user is determined, and each document is ranked. For example, when searching for a document by keyword search, the ranking is reflected in the output form of the search result so that the user can understand which document is important.

In this embodiment, the idea is that the higher the level of browsing the document, the more important the document is for the user, and the lower the level of browsing, the less important it is. The degree of importance of a document is determined based on the “degree of browsing” indicating how much interest the user has viewed with respect to the document. There are the following three methods for measuring the degree of browsing.
(A) A method of measuring the time during which content is displayed and using the measured value as browsing degree information indicating the degree of browsing.
(B) A method of measuring the number of markings made on the content and using the measured value as browsing degree information indicating the degree of browsing.
(C) A method of measuring the display frequency of content in a certain period and using the measured value as browsing degree information indicating the degree of browsing.
Hereinafter, the details of the embodiment will be described in accordance with the above concept.

(1) Configuration of Embodiment FIG. 1 is a diagram illustrating a configuration of a display search system 1 according to the present embodiment.
The display search system 1 includes a host device 10 that stores a plurality of document data (content data), a document display device 200 that displays a document based on the document data as an image, and a document displayed on the document display device 200. And a pen-type sensing device 30 for performing processing such as marking. The document data is composed of page data representing a plurality of pages constituting the document. A document ID (Identification: identification information) for identifying each document is assigned to the document data. A page number is assigned to page data included in each document data. The display retrieval system 1 can identify each document and page based on the document ID and page number, and can execute processing in units of each document / page.

  The document display device 200 includes a display unit 21, a display time measurement unit 22, and a communication unit 23. The communication unit 23 includes an antenna and a wireless communication circuit, and performs wireless communication according to a communication procedure defined by, for example, Bluetooth (registered trademark). The communication unit 23 receives document data transmitted from the host device 10 via the sensing device 30 and supplies it to the display unit 21. The display unit 21 includes a storage liquid crystal display using a cholesteric liquid crystal or electrophoresis, and a display control circuit, and displays a document based on document data supplied from the communication unit 23. The display time measuring unit 22 measures and records the time during which each document / page is displayed. This display time represents the “degree of browsing” indicating how long the user has viewed each document / page in the past. The recorded display time is transmitted to the sensing device 30 via the communication unit 23.

  The sensing device 30 includes a marking number measurement unit 31, a display frequency measurement unit 32, a storage unit 33, an image input unit 34, an image processing unit 35, and a communication unit 36. The communication unit 36 performs wireless communication according to the same communication procedure as that of the communication unit 23 of the document display device 200. The communication unit 36 receives document data transmitted from the host device 10 and stores it in the storage unit 33. The document data stored in the storage unit 33 is transmitted from the communication unit 36 to the communication unit 23 of the document display device 200 as necessary. The communication unit 36 also receives the display time transmitted from the document display device 200 and transfers it to the host device 10.

  When the user marks the document using the sensing device 30, the image input unit 34 reads an image near the marked portion. Based on the image read by the image input unit 34, the image processing unit 35 specifies where in the document the marking has been made. The marking number measuring unit 31 measures the number of markings made on the document in this way and records it for each document / page. Note that “marking” here means the content specified by the user, such as “siori”, “sticky note”, “underline” or “note” or “comment” given to the document. Information. Therefore, the number of markings represents the “degree of browsing” that indicates how much attention the user has had when browsing the document in the past. The display frequency measuring unit 32 measures and records the number of times (that is, the frequency) that a document is displayed on the document display device 200 within a certain period for each document / page. This display frequency represents the “degree of browsing” that indicates how many times the user has viewed each document in the past. Thus, the display time, the number of markings, and the display frequency all represent the degree to which the user has browsed the document, and are therefore collectively referred to as “browsing degree information”.

  The host device 10 includes a communication unit 11, a device management unit 12, a document management unit 13, a determination unit 14, a display time management unit 15, a marking number management unit 16, a display frequency management unit 17, and a storage unit. 18, a search condition specifying unit 19, and a search unit 20. The communication unit 11 performs wireless communication according to the same communication procedure as that of the communication unit 36 of the sensing device 30. The communication unit 11 transmits the document data stored in the storage unit 18 to the sensing device 30 and the browsing degree information transmitted from the sensing device 30 with the display time management unit 15, the marking number management unit 16, and the like. This is supplied to the display frequency management unit 17.

  The device management unit 12 stores a device ID assigned to the document display device 200 or the sensing device 30 in association with a document ID of a document provided to each device, and manages such registration, update, and change. Control functions. The document management unit 13 stores document IDs and page numbers of all documents, and manages management functions such as registration, update, and change. The display time management unit 15 stores the display time for each document and each page, and manages management functions such as registration, update, and change. The marking number management unit 16 stores the number of markings for each document and each page, and manages management functions such as registration, update, and change. The display frequency management unit stores the display frequency of each page and each document, and manages management functions such as registration, update, and change.

  The determination unit 14 is based on the display time stored in the display time management unit 15, the marking number stored in the marking number management unit 16, and the display frequency stored in the display frequency management unit 17. The importance (rank value) of each document and the pages included in the document is determined and stored. As schematically shown in FIG. 2, the rank value is determined for each page of each document. Further, the rank values for each page are summed and divided by the number of ranked pages. The division value (that is, the average value) becomes the document rank value.

  Returning to FIG. 1, the search condition specifying unit 19 is, for example, a keyboard provided with a plurality of operators, and receives search conditions specified by the user. The search unit 20 searches the stored content of the storage unit 18 for document data that matches the specified search condition. And the search part 20 outputs the search result in the output mode according to said rank value. Specific output means include means for displaying the search results on the document display device 200, a display device (not shown) of the host device 10, or printing out the search results on paper by a printing device (not shown). is there. Moreover, as a search method used here, a well-known thing can be used. For example, the user designates a data file name, data format, data size, creation / access date / time, or keyword, and searches for document data that matches the designated content.

(2) Rank Determination Method Here, FIG. 3 is a diagram for explaining a ranking method based on the display time, the number of markings, and the display frequency. With reference to FIG. 3, how the determination unit 14 determines the importance of each page will be described.

(2-1) Display time: t
The longer the display time, the more important the document is.
FIG. 3 shows a function T representing the relationship between the display time t and the rank value R when the horizontal axis is the display time t and the vertical axis is the rank value R. The highest rank value R is 100. As shown in FIG. 3, the rank value R increases substantially in proportion to the display time t from when the document is first displayed until a predetermined time t1. When the display time exceeds the time t1 and the rank value R reaches 100, the rank value R is maintained at 100 thereafter. That is, when the display time t1 is exceeded, the rank value R remains the highest value, no matter how long the user browses thereafter. However, for example, if the display time t reaches the time t1 and is no longer displayed, the rank value R decreases in proportion to the non-display time. It becomes. The fact that the document is no longer displayed is because it may be considered that the user's interest has decreased, that is, the importance has decreased.

(2-2) Number of markings: n
The more times a user has marked, the more important the document is.
FIG. 3 shows a function N representing the relationship between the marking number n and the rank value R, where the horizontal axis is the marking number n and the vertical axis is the rank value R. As shown in FIG. 3, up to a predetermined number n1, the rank value R increases in proportion to the marking number n. When the number of markings exceeds n1 and the rank value R reaches 100, the rank value R is maintained at 100 thereafter. That is, when the number of markings n exceeds n1, the rank value R remains the highest value no matter how many markings the user performs thereafter. Note that the rank value R based on the number of markings n does not decrease thereafter once it is increased as in the relationship between the display time t and the rank value R.

(2-3) Display frequency: f
The more frequently a document is displayed, the more important it is.
FIG. 3 shows a function F representing the relationship between the display frequency f and the rank value R when the horizontal axis is the display frequency f and the vertical axis is the rank value R. As shown in FIG. 3, the rank value R hardly increases even if the display frequency f increases until the display frequency f reaches the predetermined number of times f1. When the display frequency f exceeds the predetermined value f1, the rank value R gradually increases as the display frequency f increases. However, the increase degree of the rank value R with respect to the increase degree of the display frequency f at this time is non-linear. More specifically, until the display frequency f reaches the predetermined value f2, the function F is convex downward, and when the display frequency f exceeds the predetermined value f2, the function F is convex upward.

The reason why the function F takes such characteristics is as follows.
The stage where the user first accesses the document and displays it several times is considered to be the stage when the user starts reading the document or starts searching for a document useful for him. Therefore, at this point, it cannot be determined whether the document is necessarily important to the user. Therefore, the rank value R hardly increases until the display frequency f reaches the predetermined number of times f1. If the display frequency f further exceeds the predetermined value f1, it can be confident that the document is important for the user. Therefore, until the display frequency f reaches the predetermined value f2, the function F is convex downward and the rank value R is rapidly increased. However, it is not desirable to maintain the degree of increase of the predetermined value as it is even after the display frequency f exceeds the predetermined value f2. This is because it is not desirable to overestimate the rank value R based on the display frequency f.

For this reason, the increase degree of the rank value R with respect to the increase degree of the display frequency f is nonlinear as shown in FIG. When the display frequency f exceeds f3 and the rank value R reaches 100, the rank value R is maintained at 100 thereafter. That is, when the display frequency f exceeds f3, the rank value R remains the highest value no matter how many times the document is displayed thereafter.
As described above, the display frequency f is the number of display times during a certain period. Therefore, although not shown in FIG. 3, when the display is not so much, the display frequency f is also decreased, and the rank value R is also decreased accordingly.

(2-4) Rank calculation for document / page Next, the determination unit 14 further determines the rank value of each document / page from the rank value R determined based on the display time, the number of markings, and the display frequency. A method will be described.
here,
Rank value / t: Rank value given by the display time (0 ≦ Rank value / t ≦ 100),
Rank value / n: Rank value given by the number of markings (0 ≦ Rank value / n ≦ 100),
Rank value / f: Rank value given by display frequency (0 ≦ Rank value / f ≦ 100),
It is defined as At this time, the rank value (Rp) of each page is obtained by the following equation (i). Further, the rank value (Dp) of each document is obtained by the following equation (ii).
Page rank value (Rp) = (Rank value / t + Rank value / n + Rank value / f) / 3
(I)
Document rank value (Dp) = (ΣRp) / {Σ (number of ranked pages)} (ii)
In short, the rank value (Rp) of a page is obtained by dividing the rank value R determined based on each of display time information, marking number information, and display frequency information at a predetermined ratio. Further, the rank value (Dp) of the document is obtained by calculating a value (average value) obtained by adding the rank values (Rp) for each page and dividing the sum by the number of ranked pages.

(3) Operation of Embodiment Next, the operation of the embodiment will be described.
FIG. 4 is a flowchart illustrating a procedure executed by the determination unit 14, and FIG. 5 is a diagram illustrating an example of rank values determined by the determination unit 14 according to the method described above.
As the time for determining (calculating) the rank value, a regular time is determined in advance, for example, every 24 hours. In FIG. 4, when the determination unit 14 determines that the time for determining (calculating) the rank value has arrived (step S <b>1; Yes), it first specifies one document to be processed. Then, the display time, marking number, and display frequency corresponding to each page included in the document are read from the display time management unit 15, marking number management unit 16, and display frequency management unit 17, respectively (step S2).

Next, the determination unit 14 obtains a rank value for each page by the above-described method based on the read display time, the number of markings, and the display frequency. That is, the determination unit 14 obtains the rank value of each page according to the display time according to the functions T, N, and F shown in FIG. 3 (step S3), and obtains the rank value of each page according to the number of markings ( In step S4), the rank value of each page corresponding to the display frequency is obtained (step S5). Next, the determination unit 14 adds the rank value R of each page calculated based on each of the display time, the number of markings, and the display frequency as shown in Expression (i), and further divides by “3”. Then, a rank value (Rp) for each page is calculated (step S6). Next, as shown in the formula (ii), the determination unit 14 sums up the rank value (Rp) for each page, divides it by the number of ranked pages, and calculates the rank value (Dp) of the document. Is obtained (step S7).

  For example, in FIG. 5, ten documents from document A,..., Document J are targeted. Document A,..., Document J have “Doc1”,. “Doc10” is assigned. By calculating the rank value (Rp) of each page included in a document, the rank order of the pages in the document is specified. For example, in the example of document A, the rank value (Rp) of the page with the page number “Page9” is “90” and the rank value (Rp) of the page with the page number “Page7” from the highest rank value (Rp). ) Is “60”, the rank value (Rp) of the page with the page number “Page2” is “40”, and so on. Further, the rank value (Dp) of the document is calculated from the rank value (Rp) for each page. For example, the overall rank value (Dp) of the document A is “20”.

  The determination unit 14 repeatedly executes the processes in steps S2 to S7 as described above for all documents (step S8 in FIG. 4; No). And the determination part 14 will complete | finish the process about all the documents (step S8; Yes), will generate the ranking list | wrist which made the list from the largest rank value, and memorize | stores this (step S9).

  The ranking list generated by the determination unit 14 is divided into “by page” and “by document”. As illustrated in FIG. 5, the rank value (Rp) of the page number “Page9” of the document “Doc1” is the highest in the page-based ranking list PR, “90”. Next, the rank value (Rp) of the page number “Page3” of the document “Doc2” is “80”, and the rank value (Rp) of the page number “Page8” of the document “Doc2” is “70”. in the process of. In the document ranking list DR, the rank value (Dp) of the document “Doc3” is “90”, which is the highest. Next, the rank value (Dp) of the document “Doc2” continues at “70”, and the rank value (Dp) of the document “Doc10” continues at “30”.

  When the search condition is specified by the search condition specifying unit 19, the search unit 20 searches for document data that matches the search condition from among a plurality of document data stored in the storage unit 18. Then, when the search result is displayed or printed out, it is output in an output mode corresponding to the contents of the ranking lists DR and PR generated by the determination unit 14.

  For example, as shown in FIG. 6, it is assumed that the page numbers “Page9” of the document “Doc1” and the page numbers “Page1” and “Page8” of the document “Doc2” are extracted as search results. When the search unit 20 displays the search result at this time, the page with the page number “Page9” of “Doc1” having the highest rank value (Rp) according to the page-specific ranking list PR (here, the most prominent position (here, The page number “Page8” of “Doc2” is displayed below it, and the page of the page number “Page1” of the document “Doc2” is further displayed below it. That is, the pages included in the search result are output preferentially in descending order of the rank value (Rp).

For example, as illustrated in FIG. 7, it is assumed that a document “Doc2” and a document “Doc3” are extracted as search results. When the search unit 20 displays the search result at this time, the document of “Doc3” having the highest rank value (Dp) is displayed at the most prominent position (here, the top) according to the document ranking list DR. The document “Doc2” is displayed below the document. That is, the documents included in the search result are output preferentially in descending order of the rank value (Dp).
The user can understand at a glance which document / page is important to him / her by looking at such a search result.

  According to the above-described embodiment, the importance level of a document is determined based on the degree of past browsing for the document. Then, when searching for and outputting a document that matches a specified condition from a plurality of documents, the document is output in an output mode corresponding to the determined importance. Therefore, it is possible to present the search result in such a way that the content useful for the user can be easily understood. In addition, since the importance of each document is automatically determined by the user browsing the document, there is an advantage that the user is not burdened with special work or operation.

(4) Modification (4-1) Modification 1
The above-described embodiment may be modified as follows.
In the above embodiment, the rank value is obtained based on a plurality of types of browsing degree information such as the document display time, the number of markings, and the display frequency. However, the present invention is not limited to this method, and the rank value may be obtained based on any one or two of the browsing degree information, or the rank value may be obtained based on more browsing degree information.
In addition, as shown in equation (i), multiple types of browsing degree information are equally considered (equation (i)
In this case, the viewing degree information may be weighted instead of being divided equally by 3). For example, the display time, the display frequency, and the number of markings are respectively set to 8: 1.5: the display time is multiplied by 0.8, the display frequency is multiplied by 0.15, and the number of markings is multiplied by 0.05. You may make it consider in the ratio of 0.5. Of course, the user may arbitrarily decide how much priority is given to which browsing degree information. In short, the determination unit 14 can calculate the rank value by distributing the plurality of types of browsing degree information at a predetermined ratio.

(4-2) Modification 2
Of a plurality of types of browsing degree information such as document display time, number of markings, and display frequency, ranks may be determined based on different types of browsing degree information according to the length of the ranking target period.
The reason is that when a user views a document, there is a tendency as follows.
A sentence frequently viewed by a user only for a certain period is not necessarily a document that is important for the user in the long term. Therefore, if the browsing degree information “display frequency” is emphasized more than necessary, there is a possibility that a high rank value is given to a document that is not very important. Therefore, in the period shorter than the predetermined period, only “display time” and “number of markings” are used as criteria for rank determination, and in the period longer than the predetermined period, “display time” and “number of markings” are set to “ "Display frequency" is added, and these are used as criteria for rank determination. In this case, the determination unit 14 uses a short-term determination table as shown in FIG. 8 and a long-term determination table as shown in FIG. 9 according to the importance determination target period. For example, in the short-term determination table, the display time of Page 1 of the document with the image ID “Doc1” is “1 hour”, and the number of markings is “5”. As described above, when only the display time and the number of markings are taken into consideration without considering the display frequency, there is an advantage that the time required for the determination can be shortened because the browsing degree information as a determination reference is small.
On the other hand, in the long-term determination table, the display time of Page 1 of the document with the image ID “Doc1” is “1 hour”, the number of markings is “5”, and the display frequency is “1 time”. As described above, when all of the display time, the number of markings, and the display frequency are taken into account, since there is a large amount of browsing level information that is a criterion for determination, rank determination that more accurately reflects the browsing behavior of the user is possible.

(4-3) Modification 3
In the embodiment, no particular mention is made as to whether the number of users browsing the document is one or more. However, when there are a plurality of users, it is desirable to perform rank determination for each user. This is because which document is important depends on each user. In this case, for example, the sensing device 30 is provided with means for authenticating the user by fingerprint authentication or the like, and the user who browses the document is specified by this means. Then, the display time management unit 15, the marking number management unit 16, and the display frequency management unit 17 store the browsing degree information in association with the user identification information assigned to the user browsing the document. And when the user who browses a document is specified, the determination part 14 ranks each document and page based on the browsing degree information memorize | stored matched with the user identification information of the user. judge.

(4-4) Modification 4
The present invention is not limited to a system composed of a plurality of devices as shown in FIG. 1, but can be applied to, for example, a single personal computer, a tablet computer such as a PDA (Personal Digital Assistants), The present invention can also be applied to a single display device having high portability such as electronic paper. The program in which the procedure as shown in FIG. 4 is described is a magnetic tape, a magnetic disk, a floppy (registered trademark) disk, an optical recording medium, a magneto-optical recording medium, a CD (Compact Disk), a DVD (Digital Versatile). Disk) or a recording medium such as a RAM. That is, the present invention can be realized as a program.

It is the figure which showed the structure of the display search system which concerns on embodiment of this invention. It is the figure which showed typically the mode of ranking made in the same embodiment. It is a figure explaining the method of ranking based on display time, the number of markings, and display frequency. It is the flowchart which showed the procedure which a determination part performs. It is a figure which shows an example of the rank value which the determination part determined. It is a figure explaining the operation | movement at the time of outputting a search result. It is a figure explaining the operation | movement at the time of outputting a search result. It is a figure explaining the ranking method which concerns on a modification. It is a figure explaining the ranking method which concerns on a modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Display search system, 10 ... Host apparatus, 14 ... Determination part, 15 ... Display time management part, 16 ... Marking number management part, 17 ... Display frequency management part, 18 ... Memory | storage part, 19 ... Search condition designation part, 20 DESCRIPTION OF SYMBOLS ... Search part, 200 ... Document display device, 21 ... Display part, 22 ... Display time measurement part, 30 ... Sensing device, 31 ... Marking number measurement part, 32 ... Display frequency measurement part.

Claims (8)

Measuring means for measuring the degree of browsing of the image displayed on the display means based on the content data, and generating browsing degree information representing the degree of browsing;
Determination means for determining the importance of the content data based on the browsing degree information generated by the measurement means;
Search means for searching for content data that matches a specified search condition from among a plurality of content data, and outputting the search result in an output mode corresponding to the importance determined for each content data by the determination means A search apparatus comprising: and. The measuring means is at least one of the time when the image is displayed on the display means, the frequency at which the image is displayed on the display means, or the number of information given to the image by the user. The search device according to claim 1, wherein the search value is used as the browsing degree information. The measuring means measures the frequency at which the image is displayed on the display means, and uses the measured value as the browsing degree information,
The determination means inputs the browsing degree information generated by the measurement means for a function whose importance increases as the frequency increases, and obtains the importance of the content data,
The function is a function that protrudes downward until the frequency reaches a predetermined value when the importance is the vertical axis and the frequency is the horizontal axis, and the frequency is the predetermined value. The search device according to claim 1, wherein the search function is a function that protrudes upward when exceeding. The measuring means generates a plurality of types of browsing degree information,
2. The determination unit according to claim 1, wherein the determination unit determines the importance level of the content data by distributing the browsing levels represented by the plurality of types of browsing level information generated by the measuring unit at a predetermined ratio. The described search device. The measuring means generates a plurality of types of browsing degree information,
The determination means determines the importance level of the content data based on the browsing level information of a type corresponding to the length of a target period for determining the importance level among the multiple types of browsing level information. The search device according to claim 1. The content data is document data composed of a plurality of page data,
The measuring means measures the degree of browsing of the page displayed on the display means, and generates browsing degree information representing the degree of browsing,
The determination unit determines the importance level of each page included in each document and the importance level of each document unit based on the browsing degree information generated by the measurement unit,
The retrieval unit retrieves document data or page data that matches a designated retrieval condition from a plurality of document data stored in the storage unit, and the retrieval unit obtains each document data and each page data by the determination unit. The search device according to claim 1, wherein the output is performed in an output mode corresponding to the degree of importance determined for the page data. Comprising user specifying means for specifying a user who browses an image displayed on the display means;
The measuring means corresponds to browsing degree information indicating the degree to which the image displayed by the display means is browsed, with user identification information assigned to the user specified by the user specifying means as a user who browses the image. Remember,
2. The search according to claim 1, wherein the determination unit determines the importance of the content data based on browsing degree information stored by the measurement unit in association with each user identification information. 3. apparatus. On the computer,
A determination function for determining the importance of the content data based on browsing degree information indicating the degree to which the image displayed on the display means is browsed based on the content data;
A search function for searching content data that matches a specified search condition from a plurality of content data, and outputting the search result in an output mode corresponding to the importance determined for each content data by the determination function Program to realize and.

Images