JPH06168281A - Device and method for judging degree of importance for article information, device and method for learning keyword - Google Patents

Abstract

PURPOSE:To provide the article information importance degree judging device, which can automatically judge the importance degree of article information and can add the importance degree judged result to the article information in the case of perusing and referring to the article information by using an article information display device, and the keyword learning device which can learn preferential degree information for keywords inside a personal keyword dictionary based on the importance degree information judged by the article information importance degree judging device. CONSTITUTION:This keyword learning device is provided with an article information display means 2 for displaying the article information, article information reading time measuring means 4 for measuring time to display the specified article information by the article information display means 2, information quantity measuring means 3 for measuring the information quantity of article information, importance degree judging means 7 for judging how much the article information is important based on the article information display time measured by the article information reading time measuring means 4 and the information quantity measured by the information quantity measuring means 3, and keyword dictionary updating means for updating the priority information added to keywords inside the keyword learning dictionary based on the personal keyword dictionary storing keywords and the judgement of the importance degree judging means 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for determining the importance of article information, and a device for adding / updating priority information to a keyword in a keyword dictionary. More specifically, when browsing and referring to the article information using the article information display device, the importance of the article information is automatically determined, and an article information importance determination device and method capable of outputting the result as the importance information, Further, the present invention relates to a keyword learning device and method capable of adding priority information to a keyword entered in a keyword dictionary or learning and updating the keyword priority information based on the importance of article information.

[0002]

2. Description of the Related Art In recent years, document filing devices and the like have become increasingly large in capacity and can accommodate a large amount of documents and image information. However, even if a large amount of information can be accommodated in the document filing device or the like, if the accommodated information is not in a state where the user can efficiently use it,
The utility value of the device is halved.

Therefore, a means for efficiently searching information is indispensable for a document filing apparatus or the like, but recently, various information search systems have been proposed for the purpose of improving the efficiency of information search work and labor saving. ing. By the way, generally, when storing a document or image information in a document filing device or the like, an information retrieval system searches for information desired by a user using keyword as a clue, by inputting keyword information attached to these sentences and the like. The method is adopted. For example, in Japanese Laid-Open Patent Publication No. 1-145721, a keyword assigned to each document is attached with information indicating the degree of importance of the document, and a document close to the searcher's intention is prioritized by using the keyword with attached information. We propose a search validity judgment method for documents that is searched automatically. Further, Japanese Patent Laid-Open No. 1-123131 discloses a keyword priority automatic evaluation device which enables a search by an important keyword by extracting a keyword from individual article information and evaluating the priority of the keyword in advance. Is proposed.

[0004]

However, in the technique described in Japanese Patent Laid-Open No. 1-145721, when a desired document is browsed and referenced from a large amount of registered article information, the searcher can perform a clear search. I had to intentionally create at least one search formula and input it. Therefore, in order to efficiently search for a desired document, it is premised that the searcher has specialized knowledge about keyword selection and advanced technology for constructing a search formula.

Further, in the technique disclosed in Japanese Patent Laid-Open No. 1-131331, the keyword information given to a document and the importance of the keyword in the document are evaluated according to a certain evaluation standard irrespective of the recognition of the searching party. Therefore, a degree of importance that does not match the searcher's intention may be given. Therefore, there is a problem in that it is difficult for a searcher to truly perform a search.

In view of the above point, the present invention provides an article information importance degree determining apparatus and method capable of determining the degree of importance of article information by reflecting the degree of interest and intention of the searching party to the article content of the literature information. Another object of the present invention is to provide a keyword learning device and method capable of learning and updating the importance of keywords based on the importance of article information.

By achieving the above-mentioned object, the present invention allows the searcher to automatically search the document (article information) without special operation such as inputting a search formula. It is a further object of the invention to provide an automatic document retrieval system or the like that can preferentially and sequentially present documents of high interest.

[0008]

In order to achieve the above-mentioned object, the invention of claim 1 has an article information display means for displaying article information, and a time during which specific article information is displayed by the article information display means. Article information reading time measuring means for measuring, the information amount measuring means for measuring the information amount of the article information, the article information display time measured by the article information reading means, and the information amount measured by the information amount measuring means Based on the above, the article information importance degree determining device is provided with an importance degree determining means for determining how important the article information is.

According to the second aspect of the invention, the step of displaying article information, the step of measuring the time during which the specific article information is displayed by the article information display means, the step of measuring the information amount of the article information, Importance determination including an article information display time measured in the article information reading step and a step of determining how important the article information is based on the information amount measured in the information amount measuring step It is a method.

According to a third aspect of the present invention, in the article information importance level determining apparatus according to the first aspect, it is further monitored whether or not the operator has performed an operation other than an operation related to peripheral devices and an operation related to article information display. When the operator performs an operation other than the operation related to the peripheral device and the operation related to the article information, the time from the start of the operation to the end of the operation state based on the operation is measured, and the measured time is referred to as the article. It is characterized in that it is an importance determining device to which a device monitoring means for outputting a signal for removing from the article information display time measured by the information reading time measuring means to the article information reading time measuring means is added.

According to a fourth aspect of the present invention, in the article information importance determining method according to the second aspect, it is further monitored whether or not the operator performs an operation other than an operation related to peripheral devices and an operation related to article information display. And a step of measuring the time from the start of the operation to the end of the operating state based on the operation, when the operator performs an operation other than the operation related to the peripheral device and the operation related to the article information. And a step of outputting to the article information reading time measuring means a signal for removing the time from the article information display time measured by the article information reading time measuring means. To do.

According to a fifth aspect of the invention, in the article information importance level determination device according to the third aspect, it is further monitored whether or not the screen on which the article information is displayed is scrolled, and after the display of the article information is started. If the scroll operation is not performed before the lapse of a predetermined time, it is considered that the article information is not read by the operator, and the behavior monitoring is instructed to the importance determination means not to perform the importance determination. Means for determining the degree of importance of article information.

According to the invention of claim 6, in the article information importance judging method according to claim 4, further comprising a step of monitoring whether or not the screen on which the article information is displayed is scrolled, and displaying the article information. If the scroll operation is not performed before the lapse of a predetermined time after the start, it is considered that the article information is not read by the operator, and the importance determination means is instructed not to perform the importance determination. And an article information importance level determination method.

The invention according to claim 7 comprises article information display means for displaying article information and keyword information attached to the article information, and a personal keyword dictionary storing the keyword information. Reference is made whether or not the same keyword as the attached keyword is entered in the keyword dictionary, and if the same keyword is entered, keyword reference means for holding the keyword and the input article And a keyword dictionary updating means for updating the priority information attached to the keyword in the keyword learning dictionary based on the judgment of the importance judging means, and the importance judging means for judging how important the information is. It is a keyword learning device.

According to the invention of claim 8, the step of displaying the article information and the keyword information attached to the article information, and the same keyword as the keyword attached to the displayed article information is entered in the keyword dictionary. And a step of referring to whether or not the input article information is important, and an importance degree determining step of determining how important the input article information is,
And a step of updating the priority information attached to the keyword in the keyword learning dictionary based on the determination in the importance degree determining step.

According to a ninth aspect of the present invention, in the keyword learning device according to the seventh aspect, the importance degree determining means determines how important the article information is based on an operator's instruction. It is a device. A tenth aspect of the invention is the keyword learning method according to the eighth aspect, wherein the importance degree determining step determines how important the article information is based on an operator's instruction. Is characterized in that.

The invention of claim 11 comprises article information display means for displaying article information, and a personal keyword dictionary storing keyword information, and further, from the displayed article information,
The keyword reference unit that searches for the same keyword as the keyword entered in the personal keyword dictionary and holds the keyword, the importance determination unit that determines the importance of the displayed article information, and the keyword reference unit The keyword learning device is characterized by comprising keyword dictionary updating means for updating the priority information in the keyword learning dictionary based on the judgment of the importance degree judging means for the retrieved keyword.

According to the twelfth aspect of the present invention, an article information displaying step of displaying article information, and searching the displayed article information for the same keyword as the keyword entered in the personal keyword dictionary, and referring to the keyword holding the keyword. Step, the importance degree determining step of determining the degree of importance of the displayed article information, and the priority information in the keyword learning dictionary based on the determination of the importance degree determining step for the keyword searched in the keyword reference step. And a keyword dictionary updating step of updating.

According to a thirteenth aspect of the present invention, in the keyword learning device according to the seventh or ninth or eleventh aspect, it is possible to further enter a new keyword in the personal keyword dictionary and rewrite the priority information of the personal keyword dictionary. It is characterized in that it is a keyword learning device in which keyword information input means is added.

According to a fourteenth aspect of the present invention, in the keyword learning method according to the eighth or tenth or twelfth aspect, a step of causing a new keyword to be entered in the personal keyword dictionary and rewriting the priority information of the personal keyword dictionary is provided. It is characterized by an additional keyword learning method.

[0021]

Each structure of the invention of claim 1 operates as follows. Article information display means displays the article information. The article information reading time measuring means regards the time when the specific article information is displayed on the display by the article information displaying means as the time when the article information is read by the operator, and this time is referred to as the article information. Measure as read time. The information amount measuring means measures the information amount of the article information, for example, by measuring the number of characters. Then, the importance degree determining means determines how important the article information is based on the article information reading time measured by the article information reading means and the information amount measured by the information amount measuring means.

With each of the above-described functions acting, the invention of claim 1 determines the operator's article information reading time (apparent reading time) with respect to the input article information and the information amount of the article information. As a result, it serves to determine the importance of article information. According to a third aspect of the present invention, the article information importance degree determination device according to the first aspect is further configured by adding a device monitoring unit, and the device monitoring unit is operated by an operator regarding peripheral devices and article information. It is monitored whether an operation other than the operation related to the display is performed, and when the operator performs an operation other than the operation related to the peripheral device and the operation related to the article information, the operation state based on the operation ends from the start of the operation. Measure the time to. Then, a signal for removing the measured time from the article information reading time measured by the measuring means is output to the article information reading time measuring means. As a result, the article information importance level determination device according to claim 3 operates the peripheral device and the operations other than the operations related to the article information (periphery) while certain specific article information is displayed on the display and the display state continues. (Including the device) is performed, from when an operation other than the operation related to the peripheral device and the operation related to the article information is performed to when the operating state based on the operation ends or when the display ends. The time until the earliest is regarded as the time when the operator has not read the article information on the display, and the article information reading time can be calculated by subtracting this time from the article information display time. Therefore, the invention of claim 3 has an effect that the article information importance can be determined based on the article information reading time closer to the true value.

The invention of claim 5 is configured by further adding behavior monitoring means to the invention of claim 3, and this behavior monitoring means is predetermined after the display of certain specific article information is started. When the screen scroll operation is not performed by the operator until the time elapses, it is considered that the article information has not been read by the operator, and the importance determination unit performs the importance determination. Order not to. Therefore, according to the invention of claim 5, in addition to the operation described in claim 3, operations (including peripheral devices) other than operations related to peripheral devices and operations related to article information are performed while the display state of specific article information is continued. Even if it is not performed, if the screen scroll operation is not performed, it can be considered that the article information has not been read by the operator and can be processed, thereby avoiding meaningless importance determination. Play.

Each structure of the invention of claim 7 operates as follows. The article information display means displays article information and keyword information attached to the article information. The personal keyword dictionary stores keyword information. Further, the keyword reference means refers to whether or not the same keyword as the keyword attached to the displayed article information is entered in the keyword dictionary, and if the same keyword is entered, the keyword is referred to. Hold. The importance determining means determines how important the input article information is, based on the article information reading time. The keyword dictionary updating means updates the priority information attached to the keyword in the keyword learning dictionary based on the judgment of the importance judging means. By the above-described operations of the respective components, the invention of claim 7 is such that the keyword of the article information input together with the article information is entered when the keyword is entered in the personal keyword dictionary. The keyword learning effect of updating the importance of the keyword according to the importance of the input article information and updating the priority is lightened.

According to a ninth aspect of the present invention, in the keyword learning device according to the seventh aspect, the importance degree determining means determines how important the article information is based on an operator's instruction. As a feature, by this, importance determination can be performed by directly reflecting the operator's recognition of article information, and a keyword learning action based on this article information importance is achieved.

The respective constituent elements of the invention of claim 11 operate as follows. Article information display means displays the article information.
The personal keyword dictionary stores keyword information. The keyword reference means searches the input article information for the same keyword (character string or the like) as the keyword entered in the personal keyword dictionary, and holds the keyword. The importance determining means determines the importance of the displayed article information. The keyword dictionary updating means updates the priority information in the keyword learning dictionary based on the judgment of the importance judging means for the keyword searched by the keyword referring means. As described above, the invention according to claim 9 refers to whether or not the same keyword as the keyword entered in the personal keyword dictionary is present in the displayed article information, and if there is, holds the keyword. The keyword learning action of updating the priority information of the keyword based on the article information importance determination result made by the importance determining means is achieved.

The invention of claim 13 is the same as claim 7 or 9 above.
Alternatively, the keyword learning device described in 11 is further provided with a keyword information input means capable of entering a new keyword in the personal keyword dictionary and rewriting the priority of the personal keyword dictionary. Therefore, claim 1
In the third aspect of the invention, in addition to the operation described in claim 7 or 9 or 11, when the keyword reference means cannot retrieve the same keyword as the keyword entered in the personal keyword dictionary from the displayed article information, etc. In this case, the keyword corresponding to the article information can be newly entered in the personal keyword dictionary, and the operator can freely rewrite the priority information of the personal keyword dictionary.

Claims 2, 4, 6, 8, 10, 1
The operation of each of the method inventions 2 and 14 is basically the same as the operation of each of the inventions of claims 1, 3, 5, 7, 9, 9, 11 and 13, and therefore the description thereof will be omitted.

[0029]

EXAMPLES The present invention will be described based on examples. [First Embodiment] FIG. 1 shows a basic configuration diagram of an article information importance level judgment apparatus according to the first embodiment of the present invention. This article information importance degree determination device includes article information storage means 1, article information display means 2, information amount measurement means 3, article information reading time measurement means 4, device monitoring means 5, behavior monitoring means 6, and importance degree determination means 7. And an importance degree information output unit 8, and in terms of hardware, for example, CPU, ROM, R
It may be composed of a storage device such as an AM, a hard disk, a floppy disk, an input device such as a keyboard or a pointing device, and a display device such as a CRT or a liquid crystal display panel. Then, the operation is realized by the CPU executing the program stored in the ROM or the storage device.

Details of each of the above components are as follows.
The article information storage means 1 stores article information, and is composed of a storage device such as a floppy disk or a RAM. The article information display means 2 reads article information from the article information storage means 1 and displays it on the display. For example, a keyboard provided with a key for display instruction and a key for display end instruction and article information. It is composed of a display device for displaying. Here, the article information is not limited to a document and is used in the meaning including image information.

The information amount measuring means 3 measures the information amount of the article information displayed on the display, and the article information is measured by a known method. The article information reading time measuring means 4 measures the time (T) during which certain article information is displayed on the display by the article information display means 2.
X) is measured, and when there is a command from the device monitoring means 5, the time measured by the device monitoring means (t below) is measured.
n) is subtracted from TX, and the article information read time (T = TX
-Tn) is calculated. This article information reading time measuring means 4 has a timer as its component,
The control of the timer and the processing of the information obtained from the timer and the information output from the device monitoring means can be realized by software such as a CPU.

The device monitoring means 5 monitors whether or not the operator has performed an operation other than the operation related to the peripheral device and the operation related to the article information, and the operator performs an operation other than the operation related to the peripheral device and the operation related to the article information. In this case, the time from the start of the operation to the end of the operating state based on the operation is measured, and a signal for removing the measured time from the article information display time measured by the article information reading time measuring means is provided. It is output to the article information reading time measuring means 4.

Specifically, for example, when another window that overlaps the window in which the article information is displayed by the article information display means 2 is opened, the article information is actually read. There is a high degree of fear. Therefore, the state in which the window is opened and its time are monitored, and when a window other than the article information display is opened, it is considered that the operator has not read the article information, and the measurement time is Is output to the article information reading time measuring means 4. Also, for example, when the printer's operation panel is operated to go offline, it is highly likely that the operator has not read the article information.Therefore, monitor the time during the above operation or print operation and display the information. It is output to the article information reading time measuring means 4.

This will be described more specifically with reference to FIG. FIG. 4 is a diagram showing a state in which a plurality of windows are opened on the same screen of the display, and some of these windows overlap each other. Here, if the shaded frame A is the article information display window related to the article information storage means 1 of this device and the white frame B is another window, the article information display window A appears in the front in FIG. 4 (1). Therefore, the operator is likely to be reading the article information. On the other hand, in the case of FIG. 4 (2), since the other window B is in the foreground, the normal operator is looking at the other window B and has not read the article information. Therefore, the article information reading time measuring means 4 monitors the peripheral devices and the like, and the time (t1) when the operation related to FIG. 4 (2) is performed and the time when the state of FIG. The time (t2) is known and the information is output to the article information reading time measuring means 4. Then, the article information reading time measuring means 4 subtracts the time t (t2-t1) at which the article information was not actually read from the article information display time TX measured by itself. The monitoring at the time points t1 and t2 can be realized based on a known technique, for example, by detecting signal information input / output to / from the CPU.

The behavior monitoring means 6 monitors whether or not the screen on which the article information is displayed is scrolled, and when the scroll operation is not performed within a predetermined time after the start of the display of the article information. Means that the article information is regarded as not being read, and instructs the importance degree determination means 7 not to perform the importance degree determination. Here, the scroll includes line-by-line scrolling and page-by-page scrolling called so-called page turning.

Further, the predetermined time is determined in consideration of the reading speed of ordinary persons for article information,
In this embodiment, the settings are as follows. Let L be the maximum number of characters that the article information display unit 2 can display on one screen,
Assuming that the character reading speed of a normal person is V (the number of characters read per unit time) and the number of article information characters is C, if L <C, the article information will not fit on one screen, and the operator does not have to scroll. , Can't read the article information completely. Therefore, in such a case, if the scroll is not performed even after a sufficient time (α) has passed after the expected value [L / V] of the scrolling time interval, the operator may have removed the seat midway. It can be said that he did not read the article information. Therefore, the predetermined time is [L /
V + α], and if no scroll is made during this period, it is considered that the article information display time does not reflect the degree of interest (importance) of the operator with respect to the article information, and the importance level determination is performed. An avoidance command is output. By properly setting α, it is possible to eliminate the inconvenience that the importance determination avoidance command is erroneously issued when the amount of article information is less than the maximum amount that can be displayed on one screen and scrolling is originally unnecessary. You can also

The importance determining means 7 determines how important the article information is based on the information amount measured by the information amount measuring means 3 and the article information display time measured by the article information reading means 4. It is a judgment, and the CPU realizes it. The importance of the article information is judged by the importance judging means 7 as follows, for example. C / T is calculated based on the amount of information input from the information amount measuring means 3, for example, the number of characters C, and the article information reading time T input from the article information reading time measuring means 4. This C / T means the actual reading speed (characters / time) of the operator.

Here, since the reading speed of a normal person is about 10 characters / second, if the C / T value is larger than 10 characters / second, it can be judged that the operator has skipped the article information.
Therefore, the importance degree determination means 7 determines that the article information is not important. On the other hand, when the C / T value is less than 10 characters / second, it can be determined that the operator has read the article information. Therefore, the importance degree determination means 7 determines that the article information is important.

The judgment standard can be set freely. For example, the judgment standard is, for example, a C / T value of less than 10 has an importance of +1 and 10 to 11 has an importance of ± 0. Then, the importance degree determination means 7 can be configured to make a more detailed determination in comparison with the above. The importance degree information output unit 8 receives the result of the importance degree determination made by the importance degree determination means 7 and outputs the article information importance degree information to an external device. The basic operation of the article information importance level determining apparatus of the first embodiment configured as described above will be described with reference to the flowcharts of FIGS. 2 and 3.

First, the article information display means 2 displays one article information read from the article information storage means 1 on the display (S1). Upon receipt of this display, the article information reading time measuring means 4 resets the A timer and starts measuring the display time (S2). On the other hand, the device monitoring means 5 receives the above display and resets the B timer (S21), and at the same time, a predetermined key (including other devices) is set in advance.
Then, monitoring of whether or not the operation has been performed is started (S3). Then, when a predetermined key is turned on,
The B timer is started (S4), and it is monitored whether a predetermined key corresponding to the key is turned off (S5),
When the predetermined key is turned off before the operator performs the operation of step S7, the B timer is stopped (S6).

The A timer continues counting the time until the operator performs an operation to display the next article information or an operation to terminate the display of the article information table (S7). Also, the time measurement by the B timer is performed while the same article information is displayed on the display (S7; NO), and a certain key is O
When another predetermined key different from this key is turned on after being turned off N, the timer B is started again, and the time added to the time based on the ON / OFF of the preceding key operation is measured. . Further, in response to the display of the above-mentioned article information, the behavior monitoring means 6 starts monitoring scrolling of the display screen.

In step S7, when the operator performs an operation of displaying the next article information or an operation of ending the display of the article information table (S7; YE).
In S), the operation proceeds to step S8. Step S8
At the specified time (L
/ V + α), the article information reading time measuring means 4 stops the A timer, and the time TX is reached, provided that the display screen is scrolled (S8; YES).
Is read (S9). Further, the device monitoring means 5 stops the B timer, reads the accumulated time tn, and outputs the information to the article information reading time measuring means 4 (s1).
0). The article information reading time measuring means 4 calculates the article information reading time T = TX + tn based on TX and tn (S11). Further, the information amount measuring means 3 measures the number C of characters of the article information (S12). The information amount measuring means 3
Although the measurement of the number of characters in the article information according to step S12 is positioned in step S12 in the flowchart of FIG.
It may be performed between 1 and step S13.

In step S13, the importance degree judging means 7 calculates the importance degree judgment value C / T based on the above T and C, and in step S14 [C / T] <10 or 10≤ [C / T].
A comparison is made as to whether ≦ 11 or 11 <[C / T]. And
[C / T] <10 is “low importance”, 10 ≦ [C
/ T] ≦ 11 is “important importance”, 11 <[C /
In the case of [T], "high importance" is determined (S14 to S
17), this importance information is output as the importance of the corresponding article information (S18).

On the other hand, in step S8, before the predetermined time (L / V + α) has elapsed after the timer A was started,
When the display screen is not scrolled (S8; NO)
In this case, the operations from step S9 to step S18 are not performed, and the process directly proceeds from step S8 to step S19.
This is because if there is no scrolling within the predetermined time as described above, it is highly probable that the operator has not read the display screen, and in such a case it is decided not to perform importance determination. .

After step S18, when the operator issues a display instruction for the next article information (s19), and the next article information to be displayed remains (s20;
If NO, the step returns to S1 and the same operation as above is repeated. On the other hand, when the operator gives an instruction to end the display or when the display of all the article information ends (s20; YES), all the steps of the importance degree determination end.

The importance information output in step S18 is not limited to the usage example in the second embodiment to be described later. For example, the importance information A (important) and importance B are added to the corresponding article information.
When information such as (ordinary) or importance C (not important) is added to the storage means, and then the article information is retrieved and referenced from the storage means, a higher degree of importance is assigned. It will be provided for every conceivable use, such as enabling prioritized presentation of article information that has been
The usage mode is not the subject of the present invention, and therefore its description is omitted. . [Second Embodiment] FIG. 5 shows a basic block diagram of an embodiment of the keyword learning device of the present invention. This keyword learning device includes article information input means 100, article information display means 2, information amount measuring means 3, article information reading time measuring means 4, device monitoring means 5, behavior monitoring means 6, importance determination means 7, keyword dictionary. The update unit 101, the keyword reference unit 102, the personal keyword dictionary 103, the keyword information input unit 104, and the keyword information display unit 105 are provided, and in terms of hardware, for example, a CPU, a ROM, a RAM, a hard disk,
It can be constituted by a storage device such as a floppy disk, an input device such as a keyboard or a pointing device, and a display device such as a CRT or a liquid crystal display panel.
Then, the operation is realized by the CPU executing the program stored in the ROM or the storage device.

The details of each of the above-mentioned configurations are as follows. In the above configuration, the article information display means 2, the information amount measuring means 3, the article information reading time measuring means 4, the device monitoring means 5,
The importance degree determining means 7 is the same as that in the first embodiment, and the description thereof will be omitted. Article information input means 10
0 is for inputting article information. The article information to be input may be so-called full text itself or may be article information with keyword information already attached with keywords. FIG. 7A illustrates article information, and FIG. 7B illustrates keyword information attached to the article information.

The keyword dictionary updating means 101 updates the priority information attached to the keyword in the keyword learning dictionary based on the judgment of the importance judging means. The keyword reference means 102 refers to whether or not the same keyword as the keyword attached to the displayed article information is entered in the keyword dictionary, and if the same keyword is entered, the keyword is referred to. To hold. Here, the holding of the keyword includes, for example, a method of temporarily storing the keyword in a storage unit, a method of setting a flag for a corresponding keyword in the personal keyword dictionary 103, and the like, but the method is not particularly limited. is not.

The personal keyword dictionary 103 stores keyword information, and the stored contents reflect the individuality and intention of the operator. The keyword information input means 104 is for allowing the operator to newly enter a keyword in the personal keyword dictionary and for updating the priority information of the keyword in the keyword learning dictionary, and is composed of, for example, a keyboard or a mouse. To be done.

The keyword information display means 105 displays the keyword information (keyword and priority information attached to the keyword) in the personal keyword dictionary on the display. The basic operation of the keyword learning device of the present invention having the above components will be described with reference to the flowchart of FIG.

When article information is input through the article information input means (S101), the article information display means 2 displays one article information (S102). Next, the keyword reference unit 102 searches the article information currently displayed by the article information display unit 2 for the same keyword as the keyword entered in the personal keyword dictionary, and when the same keyword exists. The keyword is temporarily stored and retained (s103).

If the article information input in step S101 is the article information to which the keyword information is added, the keyword search by the keyword reference means will be performed for this keyword information, and on the other hand, In the case of article information to which no keyword information is attached, a character string in the article information is referred to a keyword entered in the personal keyword dictionary 103 by a known method to search for the same keyword. There is. 7A and 7B exemplify the article information to which the keyword information is attached. FIG. 7 (A) shows the article information displayed in step S102, and FIG. 7 (B) shows step S10.
1 shows the keyword information added to the article information and input.

In step S103, if the same keyword entered in the personal keyword dictionary 103 exists in the article information or the keyword information (S1)
03; YES), the operation proceeds to step S105.
For example, the keyword information attached to the article information is shown in FIG.
In FIG. 7B, if the keyword entry in the personal keyword dictionary is as shown in FIG. 7, it means that at least “car” and “car” can be learned by the keyword.

Here, step S105 is a C / T calculation procedure, the details of which are the same as the operations of steps S2 to S13 shown in FIGS. 2 and 3 of the first embodiment. In step S103, when the same keyword as the keyword entered in the personal keyword dictionary 103 is not present in the article information or the keyword information (S
103; NO), the operator proceeds to step S105 only when the operator makes an additional entry of the same keyword in the article information or the keyword information in the personal keyword dictionary 103 (S104;
Yes). On the other hand, if there is no additional entry (S10
4; NO), since there is no target (keyword) to learn the priority information, the step proceeds to S110 without performing the operation related to the keyword learning.

When the C / T calculation procedure is performed in step S105, the importance determining means 7 determines the importance of the article information based on the C / T value (S106). And C /
When T> 11, -1 is added to the importance value of the corresponding keyword in the personal keyword dictionary 103 (S107), 10≤
When C / T ≧ 11, ± 0 is added to the same keyword importance value (S108), and when C / T <10, +1 is added to the same keyword importance value (S109).

The device monitoring means 5 monitors operations performed on the keyboard of the article information display device, peripheral devices such as a printer connected to the article information display device, and the states of these devices. It is output to the article information reading time measuring means 4 as judgment information as to whether the operator is actually watching the display contents. Here, the above step S1
07 to S109 will be specifically described. For example, it is assumed that the displayed article information is FIG. 7A, the keyword information attached to the article information is the same (B), and the content of the keyword information of the personal keyword dictionary is FIG. Then, before the operation of the apparatus proceeds to step S105, an interrupt instruction operation (S1
It is assumed that an additional entry is made in the personal keyword dictionary according to 12). In this case, if the importance judgment value C / T of the article information is less than 10, “car”, “car”,
For the keyword “exhaust gas”, the priority value in the corresponding personal keyword dictionary is incremented by +1. However,
The priority value is not updated for “engine” and “motorcycle” that are not included in the keyword information attached to the article information and have not been newly added. FIG. 9 shows the contents of the personal keyword dictionary after keyword learning (FIGS. 7-9).
reference).

In the second embodiment, the keyword information display means 105 allows the contents list of the personal keyword dictionary to be viewed on the display.
While looking at the keyword information list, the operator can add / update the desired keyword or priority information to the personal keyword dictionary 103 or delete unnecessary keywords by interrupt operation at any time.

By steps S107 to S109, S1
The keyword learning operation based on one piece of article information displayed in 02 ends. Therefore, the operator performs a display instruction operation for the next article information (S110),
When there is still other article information to be displayed (S111; NO), the process returns to step S102. On the other hand, when the display end operation is performed by the operator in step S110, or when the operations of S102 to S110 are completed for all the article information input in step S101 (S111; YES), the keyword learning operation is Completed (end). It should be noted that although omitted in the flowchart of FIG. 6 to avoid complication of the flow, when keyword information is added to the input article information, as a result of the search by the keyword reference unit 102, an entry is made in the personal keyword dictionary 103. If there is a keyword that has not been added, the keyword may be automatically added to the personal keyword dictionary.

[0059]

As is apparent from the above description, according to the present invention relating to the article information importance level judging device, when the article information is browsed and referenced using the article information display device, the article information referred to by the browsing Regarding the importance, even if the operator does not pay special operation or labor, the apparatus of the present invention automatically reads the degree of interest of the operator to the article information, determines the importance of the article information, and determines the determination result as the importance. The effect that it can be added to the article information as information is obtained.

Further, according to the present invention related to the keyword learning device, even if the operator himself does not consider the keyword at all, based on the importance information reflecting the degree of interest of the operator in the article information. As a result, the priority information can be learned for the keywords in the personal keyword dictionary, and thus the personal keyword dictionary reflecting the degree of interest of the operator in the article information and the search intention can be constructed.

Further, since the article information importance judging device and the keyword learning device of the present invention do not require a complicated and large-scale mechanism as compared with the conventional device, the above effect can be realized by an economical and compact device. . As a result of the present invention having the above effects, by adding a known document search means to the personal keyword dictionary constructed by the keyword learning device of the present invention, the operator can operate from a large amount of information stored in the document filing device or the like. It is possible to automatically present article information in the priority order based on the keyword priority information that matches the operator's preference without the troublesome operations such as inputting keywords or keyword search expressions Article A search device can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram of an article information importance level determination device according to a first exemplary embodiment of the present invention.

FIG. 2 is a flowchart showing a basic operation of the article information importance level judgment device in the first exemplary embodiment of the present invention.

FIG. 3 is a flowchart showing a basic operation (continuation of FIG. 2) of the article information importance level judgment device in the first exemplary embodiment of the present invention.

FIG. 4 is a diagram showing a display state of an article information display screen.

FIG. 5 is a block diagram of a keyword learning device according to a second embodiment of the present invention.

FIG. 6 is a flowchart showing the basic operation of the keyword learning device in the second exemplary embodiment of the present invention.

FIG. 7 shows an example of article information and keyword information attached to the article information.

FIG. 8 shows the contents of a keyword dictionary before keyword learning in the second embodiment of the present invention.

FIG. 9 shows the contents of the keyword dictionary after keyword learning in the second embodiment of the present invention.

[Explanation of sign]

 1 Article Information Storage Means 2 Article Information Display Means 3 Information Amount Measuring Means 4 Article Information Reading Time Measuring Means 5 Device Monitoring Means 6 Behavior Monitoring Means 7 Importance Determining Means 8 Importance Information Output Units 100 Article Information Input Means 101 Keyword Dictionary Update Means 102 Keyword Reference Means 103 Personal Keyword Dictionary 104 Keyword Information Input Means 105 Keyword Information Display Means

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yuuki Kusumi 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (72) Tatsuya Shimoji, 1006 Kadoma, Kadoma City, Osaka Matsushita Electric Industrial Co., Ltd.

Claims (14)

[Claims] 1. An article information display means for displaying article information, an article information reading time measuring means for measuring a time during which specific article information is displayed by the article information display means, and an information amount of the article information. Information amount measuring means, article information display time measured by the article information reading means,
An article information importance degree determination device, comprising: an importance degree determination means for determining how important the article information is based on the information amount measured by the information amount measurement means. 2. A step of displaying article information, a step of measuring a time during which specific article information is displayed by the article information display means, a step of measuring an information amount of the article information, and a step of reading the article information. Article information reading time measured in step 1, and a step of determining how important the article information is, based on the information amount measured in the information amount measuring step, article information characterized by the following: Importance determination method. 3. The article information importance level determination device according to claim 1, further monitoring whether or not an operation other than an operation related to the peripheral device and an operation related to article information display is performed by the operator, and the peripheral device is determined by the operator. When an operation other than the operation related to the information and the operation related to the article information is performed, the time from the start of the operation to the end of the operating state based on the operation is measured, and the measured time is measured by the article information reading time measuring means. A device monitoring means for outputting a signal for removing from the measured article information display time to the article information reading time measuring means,
An article information importance degree judgment device characterized by being added. 4. The article information importance level determination method according to claim 2, further comprising a step of monitoring whether or not the operator has performed an operation other than an operation related to peripheral devices and an operation related to article information display; When an operation other than the operation related to the peripheral device and the operation related to the article information is performed, the step of measuring the time from the start of the operation to the end of the operation state based on the operation, and the measured time are referred to as article information reading. And a step of outputting a signal for removing from the article information display time measured by the time measuring means to the article information reading time measuring means, in addition to the step of outputting the signal. 5. The article information importance degree determination device according to claim 3, further monitoring whether or not the screen on which the article information is displayed is scrolled, until a predetermined time elapses after the display of the article information is started. If no scroll operation was performed during
Article information, characterized in that it is additionally provided with behavior monitoring means for judging that the article information has not been read by the operator and for instructing the importance degree determining means not to perform importance degree determination. Importance determination device. 6. The article information importance determining method according to claim 4, further comprising the step of monitoring whether or not the screen on which the article information is displayed has been scrolled, and a predetermined time has elapsed after starting the display of the article information. If the scroll operation is not performed before the operation, it is considered that the article information has not been read by the operator, and a step of instructing the importance degree determination means not to perform the importance degree determination is added. A method for determining the importance of article information, which is characterized by: 7. An article information display unit for displaying article information and keyword information attached to the article information, and a personal keyword dictionary storing the keyword information, and a keyword attached to the displayed article information, It is referred to whether the same keyword is entered in the keyword dictionary, and when the same keyword is entered, a keyword reference unit that holds the keyword and how important the input article information is And a keyword dictionary updating means for updating the priority information attached to the keyword in the keyword learning dictionary based on the judgment of the importance degree judging means. Keyword learning device. 8. A step of displaying article information and keyword information attached to the article information, and whether or not the same keyword as the keyword attached to the displayed article information is entered in the keyword dictionary. Based on the step of referring, and the determination in the importance determination step and the importance determination step to determine how important the input article information is, the priority information attached to the keyword in the keyword learning dictionary is displayed. And a step of updating the keyword learning method. 9. The keyword learning device according to claim 7, wherein the importance degree determining means determines how important the article information is based on an operator's instruction. apparatus. 10. The keyword learning method according to claim 8, wherein the importance degree determining step is a step of determining how important the article information is based on an operator's instruction. Method. 11. An article information display means for displaying article information and a personal keyword dictionary storing keyword information are provided, and the same keyword as the keyword entered in the personal keyword dictionary is searched from the displayed article information. A keyword reference unit that holds the keyword, an importance degree determination unit that determines the degree of importance of the displayed article information, and a keyword retrieved by the keyword reference unit based on the determination of the importance degree determination unit A keyword learning device, comprising: a keyword dictionary updating means for updating priority information in the keyword learning dictionary. 12. An article information display step of displaying article information, a keyword reference step of searching the displayed article information for the same keyword as a keyword entered in a personal keyword dictionary, and holding the keyword, the display Importance level determination step of determining the importance level of the article information, and the keyword dictionary update for updating the priority level information in the keyword learning dictionary based on the determination of the importance level determination step for the keyword searched in the keyword reference step. A method for learning a keyword, comprising: 13. The keyword learning device according to claim 7, 9 or 11, further comprising a keyword information input unit capable of further entering a new keyword in the personal keyword dictionary and rewriting the priority information of the personal keyword dictionary. A keyword learning device characterized by being added. 14. The keyword learning method according to claim 8 or 10 or 12, further comprising the step of allowing a new keyword to be entered in the personal keyword dictionary and rewriting the priority information of the personal keyword dictionary. A keyword learning method characterized by.