JPH0728831A - Graphic indicating device - Google Patents

Abstract

PURPOSE:To automatically indicate a noted graphic to which an inputted sentence refers. CONSTITUTION:The graphic inputted from a graphic input device 1 is preliminarily stored in a graphic storage file 31, plural graphic keywords composing the graphic are stored in a keyword storage file 32 and the distance of the graphics with each other is stored in a distance storage file 34 if necessary. Next, the keyword in an inputted retrieved sentence is extracted, it is stored in a keyword in sentence storage file 35, the graphic corresponding to the extracted keyword is retrieved from the graphic storage file 31 and the keyword storage file 32, and an emphasis display is performed for the graphic. The graphic is stored in a graphic number in sentence storage file 36. When plural graphics corresponding to one keyword exist, one graphic is selected from the distance storage file 34 and the graphic number in sentence storage file 36.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a graphic pointing device,
More specifically, the present invention relates to a figure designating device for automatically searching for a figure corresponding to a figure reference portion appearing in a sentence.

[0002]

2. Description of the Related Art In the case of giving a presentation using drawings, it is common practice for the presenter to explain the figures involved in the illustration by pointing with a pointer. On the other hand, in the case of displaying a figure on the display device of the computer for explanation (or for projecting it by the projector device for explanation), the movement of the icon indicating the figure in the figure is programmed in advance, and What the presenter explains in accordance with the movement of the icon is "PDS Summary # 1
0 Desktop Acc. p2 MAKI Enterprise Inc, 1990 ”.

[0003]

Although there are the conventional techniques as described above, a device for automatically pointing the figure in the figure in accordance with the presenter's explanation and the question of the questioner has not been known. Therefore, an object of the present invention is to match a presenter's explanation or a question of a questioner without requiring an operation for designating a figure and pre-programming the order of the figure to be designated. An object is to provide a figure pointing device capable of automatically pointing a figure in the figure.

[0004]

According to a first aspect of the present invention, there is provided a diagram storage means for storing a plurality of figures composing a figure by associating them with keywords, and a sentence input means for inputting a sentence. There is provided a figure designating device characterized by comprising keyword extracting means for analyzing an inputted sentence to extract a keyword and figure searching means for searching a figure corresponding to the extracted keyword from the figure storing means. . In the above configuration, the diagram storage means stores, for example, a character string included in a character string figure among a plurality of figures as a keyword in association with the character string figure. The figure includes a table. In the above configuration, the sentence input means may be, for example, a voice input device that converts voice into a sentence,
It is an OCR device that reads a character string from a printed matter.

In a second aspect, the present invention is, in addition to the above configuration, displaying a figure stored in the figure storing means and highlighting the figure searched by the figure searching means in the displayed figure. There is provided a graphic designating device characterized by further comprising display means.

In a third aspect, in addition to the above configuration, the present invention provides a distance generation for storing the distance from each figure to another figure when a plurality of figures corresponding to the same keyword are included in the figure. Means and a graphic search history storage means for storing a history of searching for a graphic by the graphic search means, and the graphic corresponding to the keyword extracted by the keyword extraction means by the graphic search means is stored in the figure storage. When a plurality of figures are stored in the means, the figure with the latest search result in the history is located at the closest distance among the figures, and the figure is used as the search result. A graphic designating device is provided.

[0007]

In the graphic designating apparatus according to the first aspect, first, a plurality of graphics forming the figure are associated with keywords and stored therein. Next, when a sentence is input, the sentence is analyzed and keywords are extracted from the sentence. And
A graphic corresponding to the extracted keyword is retrieved from the storage. Therefore, it is possible to automatically instruct a figure involved in a sentence without requiring an operation for instructing a figure and preprogramming the order of the instructed figure.

In the figure designating apparatus according to the second aspect,
The figure is displayed, and the search figure in the displayed figure is highlighted. Therefore, when giving a presentation using a figure, it is possible to automatically instruct the figure in the figure in accordance with the presenter's explanation and the question of the questioner.

In the figure designating apparatus according to the third aspect,
First, the distance from each of a plurality of figures corresponding to the same keyword to another figure is stored. Next, when a graphic corresponding to the keyword extracted from the sentence is searched, the history of the search is stored. Then, when a plurality of figure candidates are found while trying to search the figure corresponding to the keyword extracted from the sentence, among the figure candidates,
The figure having the closest geometric or logical distance is selected from the figures searched immediately before, and is selected as the search result. As a result, when explaining the arrangement order of the figures in the figure or the logical connection order of the figures, the correct figure can be automatically selected and instructed.

[0010]

EXAMPLES Examples of the present invention will be described below. The present invention is not limited to this embodiment. FIG. 1 is a diagram indicating apparatus 10 according to an embodiment of the present invention.
It is a block diagram of 0. The figure designating apparatus 100 is a CAD
A graphic input device 1 for inputting a graphic such as a device, a voice input device 2 for converting a voice into a sentence and inputting it, a storage device 3 for storing the input drawings and the like, a CPU 4, and a display device 5.
It consists of and. The storage device 3 stores a diagram analysis process 41 and a sentence analysis / graphic display process 42 as programs. In addition, the storage device 3 stores, as data, a figure storage file 31 that stores a figure composed of a plurality of input figures, and a keyword storage file that stores a set of figure numbers and keywords of a plurality of figures that form the figure. 32
And an automaton storage file 33 for storing an automaton described below, a distance storage file 34 for storing a distance described below, a sentence keyword storage file 35 for storing a keyword appearing in an input sentence, and a keyword appearing in the sentence. The in-sentence figure number storage file 36 for sequentially storing the figure numbers of the correspondingly highlighted figures is stored.

The CPU 4 executes a diagram analysis process 41 when a diagram is input and a sentence analysis / graphic display process 42 when a sentence is input. FIG. 2 is a flowchart of the diagram analysis processing 41. In step 201, a figure storing process is performed. For example, for the input of the figure group forming "Fig.48" shown in FIG. 3, figure numbers 48001-48 shown in FIG.
The data 037 is stored in the diagram storage file 31. Here, the figure storage file 31 shown in FIG. 4 will be described. The figure storage file 31 is provided with a record for each figure. The data of one record is the'figure number 'of the figure, the'figure number' which is the figure number of the figure, the'figure type 'of the figure, and the handle frame corresponding to the figure. With the upper left coordinates of the upper left coordinate of the (rectangular frame containing the figure for handling the figure as one object) and the'upper left Y value ', and the lower right coordinates of the handle frame corresponding to the figure. If a certain'bottom right X value 'and'bottom right Y value', the type of the figure is a character string, the content of the string is'character string ', and if the type of the figure is an arrow, The figure number of the figure that corresponds to the tail of the arrow, the'start figure ', and the figure number of the figure that hits the head,' end figure ', and the figures of those children when the figure is a parent and contains other figures as children inside It is composed of the figure number'included figure '. Note that, among the data in FIG. 4, nil indicates that the data is invalid.
FIG. 5 shows an example of correspondence between figures and figure numbers.

Returning to FIG. 2, in step 202, keyword extraction processing is performed. In this keyword extraction processing, the keyword corresponding to each figure is stored in the diagram storage file 31.
This is a process of taking out from the keyword storage file and storing it in the keyword storage file 32. FIG. 6 shows a flow chart of keyword retrieval processing. In step 501, the record number k indicating the order of figure numbers stored in the figure storage file 31 is set to "1". In step 502, the diagram storage file 3
The kth stored figure number 1 is searched. In step 503, the figure number nu stored in the kth position
It is determined whether the type of the figure of m is "character string". If it is "character string", the process proceeds to step 504, and if it is not "character string", the process proceeds to step 508.

In step 504, the "character string" of the graphic having the graphic number num is extracted. In step 505, it is searched whether a set (a set of a keyword and a figure number) having the “character string” as a keyword exists in the keyword storage file 32.
Here, the keyword storage file 32 will be described. As shown in FIG.
Is a set of a set of a keyword and one or more graphic numbers. It should be noted that in the keyword storage file 32, the keyword is expressed by "kana", so the above search is performed by converting the "character string" into kana. Keyword storage file 3
If no set is obtained from 2, the process proceeds to step 506. When the set is obtained from the keyword storage file 32, the process proceeds to step 507.

In step 506, the "character string" is subjected to kana conversion to be a keyword, which is paired with the figure number num and stored in the keyword storage file 32. Step 507
Then, the figure number num is added to the end of the sequence of figure numbers in the obtained set kumi to add the keyword storage file 3
Store in 2.

In step 508, it is determined whether the maximum number of graphic numbers and the record number k match. If they do not match, the process proceeds to step 509. If they match, the process ends. In step 509, "1" is added to the record number k, and the process returns to step 503.

Returning to FIG. 2, in step 203, an automaton generation process is performed. This automaton generation process is
This is a process of generating an automaton from the keyword of the keyword storage file 32 and storing it in the automaton storage file 33 as data for extracting the keyword from the sentence. FIG. 8 illustrates the automaton storage file 33. An automaton consists of nodes and links. A node number is attached to each node, and characters forming a character string are associated with each link. The automaton generation process will be described with reference to the example of FIG. 8. Each character of the character string "Kinjikisokubeesu" forming a keyword is made to correspond to a link, and a node with node number 0 is connected to a node. The end node follows the link corresponding to the last character of the keyword. Next, link each character of the character string "Kinjikisoku and Urokushosori" to the link, and connect the nodes from the node with node number 0, but since "Kinjikisoku" already exists, the subsequent nodes And branch "tookurokushori" one by one to link,
Connect the nodes and go. Next, each character of the character string "tanmatsushori" is made to correspond to a link, and the node is connected from the node of node number 0. The details of the automaton generation method are disclosed in, for example, Japanese Patent Laid-Open No. 2-71388.

Returning to FIG. 2, in step 204, distance generation processing is performed. This distance generation process is a process of generating a distance between the figures when a plurality of figures correspond to one keyword and storing the distance in the distance storage file 34. FIG. 9 shows a flow chart of the distance generation processing. In step 601, the record number k indicating the order of the figure numbers stored in the keyword storage file 32 is set to "1". In step 602, the keyword storage file 32
Get the k-th stored kumi of k. In step 603, it is determined whether the set kumi has a plurality of graphic numbers. If there is more than one, the process proceeds to step 604, and if there is only one, it is not necessary to check the distance, so the process proceeds to step 606.

In step 604, a specific figure distance set generation process is executed for each figure number forming the set kumi. A flow chart of this specific figure distance set generation processing is shown in FIG. In FIG. 10, in step 701,
Using the figure number k as an element, a distance set d (k, 0) for the figure number k and the distance 0 is generated. That is, d (k, 0) =
Let {k}. In step 702, the distance set d (k, 0)
Is stored in the distance storage file 34. This distance storage file 34 is illustrated in FIG.

In step 703, the distance number m representing the distance
Is set to "1". In step 704, for all figure numbers p that are elements of the distance set d (k, m) stored in the distance storage file 34, the distance set d (k, j Among the figure numbers which do not exist in the element of (1), all the figure numbers included in the start figure, the end figure and the inclusive figure of the record of the figure number p in the figure storage file 31 and the start figure in which the figure number p exists Alternatively, all the figure numbers having the end figure or the inclusion figure in the record are the elements of the distance set d (k, m).

In step 705, it is determined whether the distance set d (k, m) is an empty set. If it is an empty set, the process ends. If it is not an empty set, the process proceeds to step 706. Step 70
In step 6, the distance set d (k, m) is stored in the distance storage file 34. In step 707, 1 is added to the distance number m. Then, the process returns to step 704.

The above-mentioned specific figure distance set generation processing will be described in a concrete example. For example, in the generation of the distance set for the figure number '48032', first, in step 701, the distance set d.
(48032,0) = {48032} is generated and it is processed in Step 70.
2 is stored in the distance storage file 34. Next, in step 703, m = 1 is set, and in step 704, the distance set d (480
A distance set d (48032,1) is generated based on 32,0). As a figure having the element '48032' of the distance set d (48032,0) as an inclusion figure, there is a figure with a figure number 48031. Further, there is neither a figure having the element '48032' as a start figure nor a figure having an end figure. Further, the figure with the figure number 48032 has neither a start figure, an end figure, nor an inclusion figure. Therefore, only the figure number “48031” is used as an element of the distance set d (48032,1). That is, the distance set d (48032,1) = {48031}. Since the distance set d (48032,1) is not an empty set, the process proceeds from step 705 to step 706, and step 706.
Then, the distance set d (48032,1) = {48031} is stored in the distance storage file 34.

Next, in step 707, m = 2 is set, and in step 704, the distance set d (4) is set based on the distance set d (48032,1).
8032, 2) is generated. Element '4 of distance set d (48032,1)
There is no figure that has 8031 'as the included figure. Also, there is a figure number "48025" as a figure having the element "48032" as the start figure, and a figure number "48024" as the figure having the element "48032" as the end figure. Further, the graphic with the graphic number 48031 has neither a start graphic, an end graphic, or an inclusion graphic that has not been published. Therefore, figure number '4
Let 8024 and 48025 'be elements of the distance set d (48032,2). That is, the distance set d (48032,2) = {48024,480
25}. Since the distance set d (48032,2) is not an empty set, the process proceeds from step 705 to step 706, and the distance set d (48032,2) = {48024,48025} is stored in the distance storage file 34 in step 706.

Next, in step 707, m = 3 is set, and in step 704, the distance set d (4) is set based on the distance set d (48032,2).
8032, 3) is generated. Element '4 of distance set d (48032,2)
There is no figure that has 8024, 48025 'as the included figure.
Further, there is no figure having the element '48024, 48025' as the starting figure or the ending figure. Also, figure number 4
The figure 8024 is a figure number '48 that has not already appeared in the start figure.
The graphic with the graphic number 48025 has 016 'and the graphic number' 48016 'which has not been described above as the terminal graphic, but the inclusion graphic does not. Therefore, the graphic number '48016' is used as an element of the distance set d (48032,3). That is, the distance set d
(48032,3) = {48016}. Since the distance set d (48032,3) is not an empty set, steps 705 to 70
6, the distance set d (48032,3) = {4 in step 706.
8016} in the distance storage file 34.

Next, at step 707, m = 4 is set, and at step 704, the distance set d (4) is set based on the distance set d (48032,3).
8032, 4) is generated. Element '4 of distance set d (48032,3)
Graphic number 480 as a graphic having 8016 'as an inclusion graphic
There are 14 figures. In addition, the figure number “48007” is defined as a figure that has not been published and has the element “48016” as the start figure.
There is a graphic number "48008" as a graphic that has not been described and has the element "48016" as the terminal graphic. Further, the figure with the figure number 48016 has neither a start figure nor an end figure, but the figure number '4801 which has not been issued as an inclusion figure.
Holds 7 '. Therefore, figure number '48007, 48008, 4801
4,48017 'is an element of the distance set d (48032,4). That is, the distance set d (48032,4) = {48007,48008,4801
4,48017}. Since the distance set d (48032,4) is not an empty set, the process proceeds from step 705 to step 706.
In step 706, the distance set d (48032,4) = {48007,4800
8, 48014, 48017} is stored in the distance storage file 34. The same applies to m = 5 or less.

Returning to FIG. 9, in step 606, it is judged whether or not there is an unprocessed figure number.
If not, go to step 607. Step 60
7, the record number k is the keyword storage file 32.
It is determined whether or not the number matches the number of pairs stored in. If the record number k matches the number of sets stored in the keyword storage file 32, the process ends. If the record number k does not match the number of sets stored in the keyword storage file 32, the process proceeds to step 608.
In step 608, "1" is added to the record number k. Then, the process returns to step 603.

FIG. 12 is a flow chart of the sentence analysis / graphic display processing 42. In step 301, keyword extraction processing is performed. The keyword extraction process is a process of extracting a keyword included in the input sentence. In FIG.
The flowchart of a keyword extraction process is shown. Step 801
Then, it is determined whether a sentence is input from the voice input device 2. If a sentence has been input, the process proceeds to step 802, and if no sentence has been input, the process ends. Step 802
Then, the entered sentence is subjected to pattern matching using the automaton stored in the automaton storage file 33 to extract the keyword. When the keyword is obtained, the process proceeds to step 803, and when the keyword is not obtained, the process returns to step 801. The pattern matching is performed by stack initialization processing (P1) and character input processing (P1).
2), new node number storage processing (P3), node number extraction processing (P4), node number storage processing (P5), collation result output processing (P6), stack exchange processing (P7), stack number 0 and stack The two stacks of number 1 are alternately used to proceed the processing. (P1) In the stack initialization process, the stack number A is set to 0,
The stack number B is 1. In the character input process (P2), the character bj is input and output to the new node number storage process (P3). (P3) In the new node number storing process, if the node with the node number 0 can be moved by the link with the label bj, the node number of the movement destination is the stack number B.
Remember. (P4) In the node number extraction processing, the node number is extracted from the stack of stack number A. The extracted node number is deleted from the stack. (P5) The node number storing process is performed by label b from the node having the node number extracted in the node number extracting process (P4).
If the node can be moved by the link with j, the number of the destination node is stored in the stack number B. (P6) The collation result output process is performed by the new node number storage process (P3) and the node number storage process (P5).
If the possible destination node is the terminal node, the character string described as the label of the node is output as the matching result. (P7) In the stack exchange process, if the node number stored in the stack number A does not exist, the stack number A is set to 1, the stack number B is set to 0, and a new character is obtained in the character input process (P2) to obtain a new node number. The storage process (P3) and subsequent steps are executed. Details of the pattern matching are described in, for example, Japanese Patent Laid-Open No.
No. 71388.

In step 803, the obtained keywords are sequentially stored in the sentence keyword storage file 35.
FIG. 14 illustrates the in-text keyword storage file 35.

Returning to FIG. 12, in step 302, a figure number search / graphic display process is executed. This graphic number search / graphic display process is a process of searching for a graphic number from a keyword in the sentence and a process of highlighting the graphic of the searched graphic number. FIG. 15 shows a flow chart of figure number search / graphic display processing. In step 901, the keywords key are sequentially obtained from the in-text keyword storage file 35.
In step 902, the set kum having the keyword key
i is searched from the keyword storage file 32. In step 903, it is determined whether or not there are a plurality of graphic numbers num included in the set kumi. If there is only one, step 904
If there is more than one, it is necessary to select one, so the process proceeds to step 906.

In step 904, the figure number num is extracted and stored in the figure figure number storage file 36. FIG. 16 illustrates the in-text figure number storage file 36. In step 905, the graphic corresponding to the graphic number num is highlighted on the display device 5. In FIG. 17, the keyword k
An example of highlighting in the case of ey = “kinjikisokutourokushori” is shown. The handprint icon H indicates the figure number n
The figure of um = 48017 is designated as the figure of interest. The coordinate pointed by the handprint icon H is, for example, a point ¼ from the bottom of the right frame line of the handle frame of the target graphic.

In step 906, the pointer q for tracing back the history of the in-text figure number storage file 36 is set to "1". In step 907, in-text figure number storage file 3
6 is searched for the figure number num_q stored q times before. If the figure number num_q stored q times before does not exist, the process proceeds to step 908. When the figure number num_q stored q times before is obtained, the process proceeds to step 909.

At step 908, a display is made in a mode showing that the figure of interest cannot be discriminated. For example, the handprint icon H
Then, specify the blank at the upper left of the figure.

In step 909, the shortest distance figure search /
Perform graphic display processing. This shortest-distance graphic search / graphic display processing is processing for searching the graphic with the graphic number most recently designated in the past from the plurality of graphic numbers corresponding to the keyword, and highlighting the graphic. . FIG. 18 shows a flow chart of the shortest distance graphic search / graphic display processing. In step 1801, the distance number dist representing the distance is set to "1". In step 1802, the distance set d (nu
m, dist) is extracted from the distance storage file 34. If the distance set d (num, dist) is not obtained, step 180
Go to 6. If the distance set d (num, dist) is obtained, the process proceeds to step 1803. In step 1803, it is determined whether or not the figure number num_q exists in each distance set d (num, dist). If it does not exist, the process proceeds to step 1806. If it exists, the process proceeds to step 1804. In step 1804, the figure number num of the distance set d (num, dist) in which the figure number num_q existed is stored in the in-text figure number storage file 36. In step 1805, the graphic corresponding to the graphic number num is highlighted on the display device 5.

At step 1806, each figure number num
It is determined whether or not there is an unexamined distance set with respect to the distance storage file 34.
If it does not exist, the processing ends. In step 1807, “1” is added to the distance number dist, and the above step 1
Return to 802.

Shortest distance graphic search / graphic display processing 90
Set 9 with keyword key = "tanmatsushori"
umi = (tanmatsushori, 48032, 48035, 48037) is obtained, and 1 (=
q) Graphic number stored Num_num = 1 = 480
17 will be described as a specific example. In step 1801, the distance dist = 1 is set, and in step 1802, the distance set d (4
8032,1) = {48031} and distance set d (48035,1) = {4803
4} and the distance set d (48037,1) = {48036} are extracted, and these distance sets d (48032,1) = {4803 are extracted in step 1803.
1} and distance set d (48035, 1) = {48034} and distance set d (4
8037, 1) = {48036} for any figure number num_1 =
Since there is no 48017, the process proceeds to step 1806.

At step 1806, the figure number num = 48
Since 032, 48035, and 48037 have unresolved distance sets, d
Let ist = 2. In step 1802, the distance set d (480
32,2) = {48024,48025} and distance set d (48035,2) =
{48027,48028} and distance set d (48037,2) = {48029,4
8030} is taken out, and these distance set d (48032,2) = {48024,48025} and distance set d (48035,
2) = {48027,48028} and distance set d (48037,2) = {4802
Figure number num_1 = 4801 for both 9,48030}
7 does not exist, the process proceeds to step 1806.

In step 1806, the figure number num = 48
Since 032, 48035, and 48037 have unresolved distance sets, d
Let ist = 3. In step 1802, the distance set d (480
32,3) = {48016} and distance set d (48035,2) = {48020}
And distance set d (48037,2) = {48022} are extracted, and these distance set d (48032,3) = {4801 are extracted in step 1803.
6} and distance set d (48035,2) = {48020} and distance set d (4
8037, 2) = {48022} for any of the figure numbers num_1 =
Since there is no 48017, the process proceeds to step 1806.

In step 1806, the figure number num = 48
Since 032, 48035, and 48037 have unresolved distance sets, d
Let ist = 4. In step 1802, the distance set d (480
32,4) = {48007,48008,48014,48017} and distance set d
(48035,4) = {48010,48011,48013,48014,48021}
And distance set d (48037,4) = {48014,48023} are taken out, and these distance set d (48032,4) = are extracted in step 1803.
{48007,48008,48014,48017} and distance set d (48035,
4) = {48010,48011,48013,48014,48021} and distance set d (48037,4) = {48014,48023} distance set d (48032,4) = {48007,48008,48014,48017} Since the graphic number num — 1 = 48017 exists in step 1, the process proceeds to step 1804. In step 1804, the figure number '4
8032 'is stored in the figure figure number storage file 36,
In step 1805, the graphic with the graphic number “48032” is highlighted.

Returning to FIG. 15, in step 910, it is determined whether the shortest distance graphic search / graphic display process has highlighted the graphic. If highlighted, the process ends. If not highlighted, the process proceeds to step 911. Step 911
Then, "1" is added to the pointer q that traces back the history of the figure figure number storage file 36. Then, in the step 907
Return to.

[0039]

According to the figure designating apparatus of the present invention, a figure corresponding to a sentence is automatically designated. For example, when giving a presentation using figures, figures are automatically instructed in accordance with the explanation. Therefore, the operation of designating a figure becomes unnecessary while explaining. Further, even if the order of explanation is changed or there is a question, it becomes possible to automatically instruct the figure corresponding to them.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an embodiment of a graphic designating device of the present invention.

FIG. 2 is a flowchart showing a diagram analysis process.

FIG. 3 is an explanatory diagram showing an example of an input diagram.

FIG. 4 is an exemplary diagram of a diagram storage file.

FIG. 5 is an exemplary diagram showing correspondence between a figure and a figure number.

FIG. 6 is a flowchart showing a keyword extraction process.

FIG. 7 is a view showing an example of a keyword storage file.

FIG. 8 is an exemplary diagram of an automaton storage file.

FIG. 9 is a flowchart showing a distance generation process.

FIG. 10 is a flowchart showing a specific figure distance set generation process.

FIG. 11 is a view showing an example of a distance storage file.

FIG. 12 is a flowchart showing sentence analysis / graphic display processing.

FIG. 13 is a flowchart showing keyword extraction processing.

FIG. 14 is a view showing an example of a sentence keyword storage file.

FIG. 15 is a flowchart showing a graphic number search / graphic display process.

FIG. 16 is a view showing an example of an in-text figure number storage file.

FIG. 17 is an explanatory diagram of highlighted display of a figure of interest.

FIG. 18 is a flowchart showing the shortest distance graphic search / graphic display processing.

[Explanation of symbols]

 100 figure pointing device 1 figure input device 2 voice input device 3 storage device 4 CPU 5 display device 31 figure storage file 32 keyword storage file 33 automaton storage file 34 distance storage file 35 sentence keyword storage file 36 sentence figure number storage file 41 diagram analysis Processing 42 Sentence analysis / graphic display processing 201 Figure storage processing 202 Keyword extraction processing 203 Automaton generation processing 204 Distance generation processing 301 Keyword extraction processing 302 Graphic number search / graphic display processing 909 Shortest distance search / graphic display processing

Claims (5)

[Claims] 1. A diagram storage means for associating a plurality of figures constituting a figure with keywords to store them, a sentence inputting means for inputting a sentence, and a keyword extraction for analyzing the inputted sentence to extract a keyword. A graphic designating device comprising means and graphic search means for searching the graphic storage means for a graphic corresponding to the extracted keyword. 2. The figure designating apparatus according to claim 1, wherein the figure storing means stores a character string included in a character string figure among a plurality of figures as a keyword in association with the character string figure. A graphic designating device characterized by: 3. The graphic pointing device according to claim 1, wherein the sentence input means is a voice input device for converting a voice into a sentence. 4. The graphic designating apparatus according to claim 1, wherein the graphic stored in the graphic storage means is displayed, and the graphic searched by the graphic search means is highlighted in the displayed graphic. A graphic designating apparatus further comprising graphic highlighting means for displaying. 5. The figure designating device according to claim 1, wherein when a plurality of figures corresponding to the same keyword are included in the figure, the distance from each figure to another figure is determined. The method further comprises distance storing means for storing, and graphic search history storing means for storing a history of searching for a graphic by the graphic searching means, and the graphic searching means has a graphic corresponding to the keyword extracted by the keyword extracting means. Is stored in the figure storage means, a figure having the closest search result in the history is selected from those figures, and the figure is searched. A graphic designating device characterized by a result.