JP2009070299A - Document processor, program, and document processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a document processor capable of contributing to the improvement of business efficiency by reducing labor of a user, and to provide a program, and a document processing method. <P>SOLUTION: An event information extracting part 12 extracts event information from an accessed electronic document, and a task predicting part 13 predicts existence of relation to an existing task. When relation to the existing task is judged, a task relating part 14 performs relating to the existing task, to update the registration document of the existing task. Consequently, it is predicted that the document is not related to any existing tasks. Even in this case, the new task is easily generated without allowing the user neither to read the content of the document nor to take time, labor and cost. Then the event information of the task is compensated, so that the document can be related to the new task. Thus, the labor of the user is further reduced compared with the conventional manner, and contribution to the improvement of business efficiency is attained. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a document processing apparatus, a program, and a document processing method.

  A task, which is a minimum unit of work for achieving an objective, may be accompanied by a task name, due date, priority, and the like. Task management refers to grasping task due dates, priorities, execution / non-execution, and the like. By performing task management, it is possible to grasp the amount of work, improve work efficiency, and check work priorities. In addition, by managing tasks, it is possible to reduce mistakes and avoid risks. For example, in the information management system disclosed in Patent Document 1, information related to work such as a schedule, an electronic mail document, and a report is managed by a task.

  Conventionally, task registration and association between tasks and various documents have been performed manually. According to the information management system of Patent Document 1, task registration and task selection are left to the user. More specifically, as described above, when a user registers a document in the information management system, the user has to determine which task the document belongs to or a new task based on the content of the document. Don't be. Such a problem requires labor and labor as the number of documents and tasks increases in document sorting and association and task registration. This is particularly noticeable when someone other than the document creator performs this task.

  Therefore, Patent Document 2 proposes a technique for automatically assigning various documents to tasks by a rule-based method using a dictionary for determining whether a target document belongs to one or more specific categories. ing. More specifically, according to the technique disclosed in Patent Document 2, a dictionary describing the relationship (rule) between tasks and documents is used. For example, in Patent Document 2, a classification category corresponding to an attribute value entry in a document “information collection / classification for hypothesis formation” “analysis of the information” “generation and verification of hypothesis” “knowledge based on the verification result” Classification rules describing "sharing of". In addition, there is a method of setting a keyword and applying a dictionary in which the degree of matching of the keyword with the distribution target document is made regular.

JP 2006-099581 A JP 2006-066558 A

  However, according to the technique disclosed in Patent Document 2, there is a problem that if an attribute value or a keyword in a document is not included in a dictionary entry, the document is not classified. In such a case, it is necessary for the user to make a judgment when it is not classified.

  The present invention has been made in view of the above, and an object of the present invention is to provide a document processing apparatus, a program, and a document processing method that can reduce the user's labor and contribute to the improvement of business efficiency.

  In order to solve the above-described problem and achieve the object, the invention according to claim 1 is an event information extraction unit that extracts event information representing an event from the text document in a document processing apparatus that processes the text document; Task storage means for storing event information of an existing task, and for the event information extracted by the event information extraction means, an existing task is stored using the event information of the existing task stored in the task storage means. A task prediction unit that predicts the association of the text document and the text document having the event information predicted to be associated with the existing task by the task prediction unit, and the task storage Task association means for updating a task stored in the means.

  According to a second aspect of the present invention, in the document processing device according to the first aspect, when it is predicted that the task prediction unit has no relation to an existing task, a new task is generated using event information of the text document. And a task registration unit for registering a task newly generated by the task generation unit in the task storage unit together with event information of the text document.

  According to a third aspect of the present invention, in the document processing apparatus according to the second aspect, the task generation means is acquired from the text document having the event information predicted to be unrelated to an existing task. The new task name and the event information of the new task are generated using the event information phrases.

  According to a fourth aspect of the present invention, in the document processing apparatus according to the second aspect, the task registration unit sets a new registration flag for a task newly generated by the task generation unit.

  The invention according to claim 5 is the document processing device according to any one of claims 1 to 4, wherein the event information extraction means clearly indicates the relationship of the structure obtained by decomposing the text document from the viewpoint of grammar. The event information is extracted using an integrated structure.

  The invention according to claim 6 is the document processing apparatus according to claim 5, wherein the event information extraction unit has a machine-readable data format, and the 5W1H and the predicate related information that are related to each other in the text document. Are extracted as the event information.

  According to a seventh aspect of the present invention, in the document processing apparatus according to any one of the first to sixth aspects, the task prediction means stores the event information of the text document in the existing task storage means. Task suitability judging means for judging suitability of the task using event information of the task, and the task predicting means is related to an existing task based on the suitability judgment by the task suitability judging means. Predict.

  The invention according to claim 8 is the document processing apparatus according to any one of claims 1 to 7, further comprising document access means for accessing the digitized document, wherein the event information extraction means is included in the digitized document. Load text documents.

  According to a ninth aspect of the present invention, there is provided the document processing apparatus according to any one of the first to seventh aspects, wherein the scanning document reading device reads the paper document via the image reading device, and the paper document read by the scanning scanning device. Character recognition means for acquiring a text document in the image data, and the event information extraction means reads the text document in the image data.

  The invention according to claim 10 is a program for causing a computer to execute document processing for processing a text document, and an event information extraction function for extracting event information representing an event from the text document, and an event of an existing task A task storage function for storing information in a storage unit, and an existing task using event information of an existing task stored in the storage unit by the task storage function for the event information extracted by the event information extraction function A task prediction function for predicting the association with the task, and the text document having the event information predicted to be associated with the existing task by the task prediction function is associated with the corresponding existing task, and the task A task association function for updating a task stored in the storage unit with a storage function; To be executed by the data.

  Further, in the invention according to claim 11, in the program according to claim 10, when it is predicted that the task prediction function is not related to an existing task, a new task is generated using the event information of the text document And a task registration function for registering a task newly generated by the task generation function in the storage unit together with event information of the text document.

  According to a twelfth aspect of the present invention, in the program according to the eleventh aspect, the task generation function is an event acquired from the text document having the event information predicted to be unrelated to an existing task. A new task name and event information of the new task are generated using the information phrase.

  The invention according to claim 13 is the program according to claim 11, wherein the task registration function sets a new registration flag for a task newly generated by the task generation function.

  Further, the invention according to claim 14 is the program according to any one of claims 10 to 13, wherein the event information extraction function is an integrated structure in which the relationship of the structure in which the text document is decomposed from the viewpoint of grammar is specified. The event information is extracted using.

  The invention according to claim 15 is the program according to claim 14, wherein the event information extraction function is a machine-readable data format, and the 5W1H and the predicate related information in the text document are related to each other. Extracted as the event information.

  According to a sixteenth aspect of the present invention, in the program according to any one of the tenth to fifteenth aspects, the task prediction function is configured to store existing task information stored in the storage unit with respect to event information of the text document. It includes a task suitability determination function that determines task suitability using event information, and the task prediction function predicts an association with an existing task based on the suitability determination by the task suitability determination function.

  According to a seventeenth aspect of the present invention, in the program according to any one of the tenth to sixteenth aspects, the computer further executes a document access function for accessing the digitized document, and the event information extraction function includes the electronic information document. A text document in a document is read.

  According to an eighteenth aspect of the present invention, in the program according to any one of the tenth to sixteenth aspects, a read scanning function for reading a paper document via an image reading means, and an image of the paper document read by the read scanning function. A character recognition function for acquiring a text document in the data is further executed by the computer, and the event information extraction function reads the text document in the image data.

  The invention according to claim 19 is a document processing method in a document processing apparatus for processing a text document, wherein an event information extraction step for extracting event information representing an event from the text document, and an existing task A task storage step for storing the event information in the storage unit, and the event information extracted in the event information extraction step, using the event information of the existing task stored in the storage unit in the task storage step A task prediction step for predicting the association with the task, and the text document having the event information predicted to be related to the existing task in the task prediction step, and the corresponding existing task, A task association step of updating the task stored in the storage unit in the task storage step.

  According to the present invention, event information is extracted from the accessed electronic document, the presence / absence of association with an existing task is predicted, and when it is determined to be associated with an existing task, the existing task is associated with the existing task. By updating the registered document, it is possible to predict that the document is not related to any existing task, and even in this case, it is very easy to perform a new task without reading the contents of the document and without the effort and cost. It is possible to generate and supplement the event information of the task and associate the document with the new task, so that the user's labor is greatly reduced compared to the conventional, and it can contribute to the improvement of business efficiency. There is an effect.

  Exemplary embodiments of a document processing apparatus, a program, and a document processing method according to the present invention will be explained below in detail with reference to the accompanying drawings.

[First Embodiment]
A first embodiment of the present invention will be described with reference to FIGS. This embodiment is an example in which a server computer is applied as a document processing apparatus.

  FIG. 1 is a module configuration diagram of a document processing apparatus 1 according to the first embodiment of the present invention. The document processing apparatus 1 is, for example, a general personal computer.

  As shown in FIG. 1, a document processing apparatus 1 includes a CPU (Central Processing Unit) 101 that performs information processing, a ROM (Read Only Memory) 102 that is a read-only memory storing a BIOS, and various data that can be rewritten. Random Access Memory (RAM) 103, HDD (Hard Disk Drive) 104, which functions as various databases and stores various programs, and stores information using a storage medium 110 or distributes information to the outside. A medium drive device 105 such as a CD-ROM drive for obtaining information from the outside, a communication control device 106 for transmitting information by communication with a server connected to an intranet or the Internet via the network 2, CRT (Cathode Ray Tube) or LCD (Liquid Crystal Disp) that displays processing progress and results to the operator lay) and the like, and an input unit 108 such as a keyboard and a mouse for an operator to input commands and information to the CPU 101. Data transmitted and received between these units is a bus controller. 109 operates in arbitration.

  In such a document processing apparatus 1, when the user turns on the power, the CPU 101 activates a program called a loader in the ROM 102, and loads a program for managing computer hardware and software called an OS (Operating System) from the HDD 104 into the RAM 103. This OS is started. Such an OS activates a program, reads information, and stores information in accordance with a user operation. As a representative OS, Windows (registered trademark) and the like are known. These operation programs running on the OS are called application programs. The application program is not limited to one that runs on a predetermined OS, and may be one that causes the OS to execute some of the various processes described below, or constitutes predetermined application software, an OS, or the like. It may be included as part of a group of program files.

  Here, the document processing apparatus 1 stores a document processing program in the HDD 104 as an application program. In this sense, the HDD 104 functions as a storage medium that stores the document processing program.

  In general, application programs installed in the HDD 104 of the document processing apparatus 1 include various optical disks such as CD-ROM and DVD, various magnetic disks such as various magneto-optical disks and flexible disks, and various kinds of semiconductor memories. The operation program recorded on the storage medium 110 such as a system medium is recorded on the storage medium 110 and installed in the HDD 104. Therefore, the portable storage medium 110 such as an optical information recording medium such as a CD-ROM or a magnetic medium such as an FD can also be a storage medium for storing an application program. Furthermore, the application program may be imported from the outside via the communication control device 106 and installed in the HDD 104, for example.

  In the document processing apparatus 1, when a document processing program operating on the OS is started, the CPU 101 executes various arithmetic processes according to the document processing program and centrally controls each unit. Of the various types of arithmetic processing executed by the CPU 101 of the document processing apparatus 1, characteristic processing of the present embodiment will be described below.

  FIG. 2 is a block diagram illustrating a schematic configuration of the document processing apparatus 1. As shown in FIG. 2, the document processing apparatus 1 follows a document processing program, whereby a document access unit 11 that functions as a document access unit, an event information extraction unit 12 that functions as an event information extraction unit, and a task prediction unit. A task prediction unit 13 that functions, a task association unit 14 that functions as a task association unit, a task generation unit 15 that functions as a task generation unit, a task registration unit 16 that functions as a task registration unit, an information storage unit 17, Is provided.

  Also, the document processing apparatus 1 follows the document processing program to store the document list 21 for storing the digitized document, the event list 22 for storing the event information, and the event information of the existing task in the HDD 104 that is the storage unit. And a task DB 23 functioning as a task storage means for storing.

  Next, each part will be described.

  The document access unit 11 is an electronic document stored in a system having a document database that can be accessed using the network 2, such as a mail server 3, a document server 4, a bulletin board server 5, and a scheduler 6 via, for example, the Internet or an intranet. The group is accessed and stored in the document list DB 21. The document access unit 11 assigns a document ID to the accessed electronic document. FIG. 3 is an example of an accessed electronic document.

  The event information extraction unit 12 reads the digitized document accessed by the document access unit 11, extracts all event information that can be acquired in the digitized document, and stores it in the event list 22.

  Here, the event information extraction process from the digitized document will be described with reference to the flowchart of FIG. As shown in FIG. 4, in the event extraction process, when there is a text in the digitized document accessed by the document access unit 11 (Yes in step S1), the text in the digitized document is read (step S2). Is obtained (step S3). The syntactic structure is a structure in which the structure of a sentence is disassembled from the viewpoint of grammar and its relationship is clarified, and an existing natural language analysis technique is used to acquire the syntactic structure. FIG. 5 is an example of syntactic structure extraction from the text of the example document of FIG. As shown in FIG. 5, the syntactic structure is composed of a phrase notation that constitutes a sentence, a phrase part of speech that is a grammatical attribute of the phrase, and a dependency relation between the phrase and the phrase (relation destination number). In addition to the configuration shown in FIG. 5, there may be information such as word information constituting a clause, dependency relationship name, and the like. The syntactic structure expression method may be a graph structure or a tree structure in addition to the structure shown in the right part of FIG.

  When the syntactic structure is obtained as described above, event information representing an event is extracted using the syntactic structure of the sentence in the subsequent step S4. In the event information extraction, event information including each element of 5W1H (When = when, Where = where, Who = who, What = what, Why = why, How = (how and how)) + predicate is extracted. In this embodiment, a knowledge dictionary describing knowledge using grammatical characteristics is referred to. FIG. 6 is a schematic diagram illustrating an example of description in the knowledge dictionary. The knowledge dictionary is a dictionary in which at least a sentence syntactic structure, corresponding 5W1H event attributes, a headword phrase and each event attribute are described. Further, as a constituent element of the dictionary, a further semantic attribute may be added to the semantic interpretation of 5W1H. In FIG. 6, detailed semantic attributes are given to When information and Where information such as “range base point” and “range end point”. In addition to the description method shown in FIG. 6, the description may be in a form in which grammatical information is expressed in regular expressions. In the case of a system implementation, it is appropriate to use a rule in which such knowledge is written down in the form of a program code. . The event information extraction unit 12 first searches for a predicate clause from the end of the syntactic structure of the first sentence, and uses a knowledge dictionary to determine whether the elements of the clause having a dependency relationship with the predicate clause can be event information for each predicate clause. If event information is determined, an event ID is added to the predicate clause and the corresponding clause, and an event attribute and an event element are registered in the event list together with the document ID and event ID. If there is another predicate clause in one document and there is an event element, the event information extraction unit 12 repeats the event information extraction until there is no event information. Furthermore, if there is a headword at the itemized part in the text, the event information extraction unit 12 uses the headword as an event attribute and extracts the element as an event element.

  When event information is extracted (Yes in step S5), an event ID is added (step S6), and the event information is registered in the event list 22 together with the event information, document ID, and event ID (step S7).

  Such processing in steps S4 to S7 is repeated until there is no event information in the digitized document accessed by the document access unit 11 (No in step S8).

  If event information is not extracted from the digitized document accessed by the document access unit 11 (No in step S5), an error display is displayed on the display unit 107 (step S9), and the process ends.

  FIG. 7 shows an example of event information extracted from the syntactic structure of the text of FIG. 5 using the knowledge dictionary of FIG. Each event information includes a document ID, an event ID, an event attribute such as When and Who, and a phrase notation or a phrase notation sequence that is an event element.

  Here, an example of extracting event information in FIG. 7 will be described in detail. First, the document ID is detected and stored. Next, the text portion at the beginning of the text and enclosed in corner brackets is extracted as a document type and stored. Next, a date clause that is right-justified and has a different character size is interpreted as a document issuance date, and an organization / location clause that is also right-justified and has a different character size is interpreted as an issuer. In the present embodiment, these are not stored, but may be stored in advance.

  Next, predicate clauses are searched in order from the first sentence. First, the predicate “I will become” is found, and the content word is converted into the final form “is” and stored. The clause related to the predicate clause is searched, and “establishment LAN is” and “disconnect” are found. Referring to the knowledge dictionary in FIG. 6, “established LAN” has a What attribute because the content word is a noun and is a case-qualified clause, and “disconnected” has the content word as an adjective verb In addition, since it is a dignity modification clause, it is determined that it has a How attribute, and these are stored as event information together with the event ID.

  Search for the next predicate clause, find "Please note", convert the content word to the final form "Caution" and store it. The clause related to this predicate clause is searched, and “related party” is found. With reference to the knowledge dictionary, it is determined that “the person concerned” does not have an event attribute, and this phrase relation is not event information.

  By the way, a predicate does not appear after the next sentence, but since there are numbered independent clauses, it is determined that a bullet is present, and the bulleted headword is used as an event attribute.

  First, since “1. reason” matches the entry word {reason} in the knowledge dictionary in FIG. 6, the event element “Why” is set as the event element “whether the network is stopped due to the electrical equipment forensic inspection” as the event element. Remember. Next, since “2. Date and time” matches the headword {date}, the event element “When” has the following elements “Saturday, June 24” “8:00 to 14:00” “(Japan time) As an event element. In the embodiment, the breakdown attribute of When is divided into when-date, when-time-range, when-other, and the like by using a knowledge dictionary, and the elements are subdivided and stored. May be "Saturday, June 24, 8:00 to 14:00 (Japan time)". Next, since “3. Range” matches the headword {time: range} or the headword {location: range}, and the subsequent element includes a content word indicating the location, the headword {location: range} Specify the event attribute as “Where-range” and the subsequent element “all sections of the Yokohama office” as the event element. Furthermore, since “4. inquiry destination” matches the headword {inquiry destination}, the subsequent element “XX Systems Co., Ltd.” “Tel: 045-111-2222” is used as the event element as the event attribute Whom. . In the embodiment, the breakdown attribute of Whom is divided into Whom and Whom-Tel using the knowledge dictionary, and the elements are divided and stored. However, the elements are collectively referred to as “XX Systems, Inc. Tel: 045-111-2222 ".

  As a result, event information with event attributes of event IDs 001 to 005 as shown in the event information example on the right side of FIG. 7 can be acquired from the syntactic structure example on the left side of FIG. In the processing flow of FIG. 4, this part is from the event information extraction process (step S4) to the end.

  The task prediction unit 13 performs task prediction when all event information can be extracted from one document. More specifically, the task prediction unit 13 includes a task suitability determination unit 13a that functions as a task suitability determination unit, and the event information of the target electronic document is stored in the task DB 23. The suitability of the task is determined using the event information of the task, the task of the target digitized document is predicted, and the subsequent processing is distributed. If there is a matching event element, the subsequent process is distributed by assigning matching information to the event element. Note that a plurality of task prediction results may be displayed on the display unit 107, and the user may select a desired task via the input unit.

  The task association unit 14 associates the computerized document predicted to be related to the existing task by the task prediction unit 13 with the corresponding existing task, and stores the task stored in the task DB 23 via the information storage unit 17. Update.

  The task generation unit 15 generates a new task using the event information of the digitized document for the digitized document predicted to be unrelated to the existing task by the task prediction unit 13.

  The task registration unit 16 registers the task newly generated by the task generation unit 15 together with the event information, and updates the task stored in the task DB 23 via the information storage unit 17.

  Here, the flow of task prediction, task association, and task new registration processing will be described with reference to the flowchart of FIG. As shown in FIG. 8, when it is determined that there is registration of new event information extracted by the event information extraction unit 12 (Yes in step S11), event information (document ID, event ID, event attribute, event Element) is read and stored in the buffer (step S12). Next, the matching process with the event element for each event attribute of each task of the existing task is executed (step S13), and if there is a matching event element (Yes in step S14), the event element of the event list 22 includes the matching event element. The matching information is given and stored in the buffer (step S15). Subsequently, event information compatibility is determined according to the number of matching elements between event elements (step S16). For example, there is a method of quantifying the number of matching elements among all event elements, and evaluating and judging using, for example, a TF-IDF index. If there is no matching event element (No in step S14), the process proceeds to step S16. If it is determined in the suitability determination in step S16 that the existing task is related (Yes in step S17), the document ID is added to the target task (step S18). On the other hand, when it is determined in the suitability determination in step S16 that the existing task is not related (No in step S17), a new task is generated (step S19), and each event element of the new task is assigned to the electronic element. Registered and stored in the event element of the document (step S20). Then, as described above, when a new task is generated and stored and a document ID is added to an existing task, the task in the task DB 23 is updated and an update flag is assigned to the task in the task DB 23 (step S21). By doing so, the update information can be notified to the user at the time of task output. The processes in steps S12 to S21 as described above are repeated until there are no registered events (No in step S22).

  FIG. 9 is a schematic diagram illustrating an example of task association processing. The task list in the task DB 23 shown in FIG. 9 stores event information (document ID, event ID, event attribute, event element) associated with each task with a task ID. Here, a detailed description will be given of a case where event information of the electronic document (document ID 001) shown in FIG. 3 is newly registered. When the event information of the digitized document (document ID 001) shown in FIG. 3 is newly registered as an event, matching processing is performed with each element for each event attribute of each task. The event information shown in FIG. 9 is different from the event information shown in FIG. 7 in that detailed elements are grouped and the Who attribute having no event element expresses that there is no event element.

  The determination of the compatibility between the event information of the electronic document to be processed and the event information of each existing task stored in the task list of the task DB 23 is as follows.

  Specifically, first, it is determined whether or not the predicate attribute element of the event information of the task ID 001 in the task list matches the predicate attribute element of the new event information. As shown in FIG. 9, since “is” matches in the task ID 001, the matching information is given to the predicate attribute. Similarly, it is determined whether the What attribute element of the task ID 001 event information matches the What attribute element of the new event information. Since the “office LAN” matches, the conformance information is added to the What attribute. Similarly, it is determined whether the Who attribute element of the event information of the task ID 001 matches the Who attribute element of the new event information. Since there is no event element, the matching information is given to the Who attribute. Similarly, it is determined whether the element of the event attribute of task ID 001 matches the element of the How attribute and the element of the How attribute of the new event information. Since “not connected” matches, the matching information is given to the How attribute. Similarly, it is determined whether the Why attribute element of the event information of the task ID 001 matches the Why attribute element of the new event information. Since “stop of network due to electrical equipment court inspection” coincides, conformity information is given to the Why attribute. Similarly, it is determined whether or not the attribute of the “When” attribute of the event information of the task ID 001 matches the element of the “when” attribute of new event information. Here, since the elements do not match, it is determined that they do not match and nothing is given. Further, it is determined whether the Where attribute element of the event information of the task ID 001 matches the Where attribute element of the new event information. Since “all sections of the Yokohama office” match, the conformance information is given to the Where attribute. Since there is no event information of the task ID 001, it is determined that the Whom attribute of the new event information does not match, and nothing is given. Then, the degree of matching between the new event information and the event information of the task ID 001 for each attribute is determined, and the degree of matching between the new event information and the event information of the task ID 001 is high. Therefore, the document ID 001 having the new event information Is determined to be related to the existing task ID 001. If it is determined that there is a relationship, the document ID 001 (indicated by shading in FIG. 9) of the digitized document shown in FIG. 3 is registered in the document ID section of the task ID 001 as a task prediction candidate, and each event information is also existing What attribute element “June 24 (Sat)” and Who attribute “XX Systems, Inc., 045-111-2222” information (not shown in FIG. 9) The information of the task ID 001 is updated. At that time, an update flag may be added.

  Next, it is determined whether the predicate attribute element of the event information of task ID 002 in the task list matches the predicate attribute element of the new event information. As shown in FIG. 9, since the elements do not match in the task ID 002, it is determined that they do not match and nothing is given. Subsequently, it is determined whether or not the What attribute element of the event information of the task ID 002 matches the What attribute element of the new event information. Since the elements do not match, it is determined that they do not match and nothing is given. Similarly, the matching of the elements of the Who attribute, the where attribute, the Where attribute, the Why attribute, the How attribute, and the Who attribute is determined. Since none of the elements match, it is determined that the document ID 001 having this new event information is not related to the existing task ID 002.

  Similarly, the predicate attribute element of the event information of task ID 003 in the task list is matched with the predicate attribute element of the new event information. As shown in FIG. 9, since no element matches in task ID 003, it is determined that document ID 001 having this new event information is not related to existing task ID 003.

  In this way, this suitability determination is performed on all existing tasks in the task list, and it is determined that there is a relationship with a task having a high degree of conformity. The document ID 001 of the document in FIG. 3 is registered, necessary information is added and stored, and the information of the corresponding task is updated.

  When task prediction and association to existing tasks are completed for all newly registered documents, the task prediction result is displayed for each newly registered document on the display unit 107, and the user selects and determines a task using the input unit 108 or the like. May be. Of course, all the prediction results may be automatically associated with the task.

  On the other hand, FIG. 10 is a schematic diagram showing an example of a task new registration process. The task list in the task DB 23 shown in FIG. 10 stores event information (document ID, event ID, event attribute, event element) associated with each task with a task ID. Here, a case where event information of a new electronic document (document ID 002) is newly registered will be described in detail.

  Judgment of compatibility between the event information of the target electronic document and the event information of each existing task stored in the task list of the task DB 23 is as follows.

  Specifically, first, it is determined whether or not the predicate attribute element of the event information of the task ID 001 in the task list matches the predicate attribute element of the new event information. As shown in FIG. 10, since the elements do not match, it is determined that they do not match, and nothing is given. Next, it is determined whether the What attribute element matches the What attribute element of the new event information. Since the elements do not match, it is determined that they do not match and nothing is given. Similarly, the matching of the elements of the Who attribute, the where attribute, the Where attribute, the Why attribute, the How attribute, and the Who attribute is determined. Since none of the elements match, it is determined that the document ID 002 having this new event information is not related to the existing task ID 002.

  Next, it is determined whether the predicate attribute element of the event information of task ID 002 in the task list matches the predicate attribute element of the new event information. Since the elements do not match, it is determined that they do not match and nothing is given. Subsequently, it is determined whether or not the What attribute element of the event information of the task ID 002 matches the What attribute element of the new event information. Since the elements do not match, it is determined that they do not match and nothing is given. Similarly, the matching of the elements of the Who attribute, the where attribute, the Where attribute, the Why attribute, the How attribute, and the Who attribute is determined. Since none of the elements match, it is determined that the document ID 002 having this new event information is not related to the existing task ID 002. Similarly, the compatibility determination is performed for the existing task ID 003 and it is determined that there is no relation.

  Thus, this suitability determination is performed for all existing tasks in the task list, and if it is predicted that there is no relation to any existing task, a new task is generated. Specifically, for example, an ID that does not conflict with an existing task is prepared (in FIG. 10, a new task ID 004), and the element of the new event information is entered as it is in the event attribute element of the task. Then, the document ID 002 of the new document is registered in the section of the document ID of the new task as a document related to the new task, stored in the task, and the task is updated. When there is a new task generation, by setting a new registration flag, the output unit can be designed to deliver new registration information to the user when the task is updated.

  As described above, according to the present embodiment, when event information is extracted from the accessed electronic document, the presence / absence of the association with the existing task is predicted, and the existing task is determined to be related. And update the registered document of the existing task, it can be predicted that the document is not related to any existing task, and even in this case, the user does not read the contents of the document, and without much effort and cost Since it is possible to easily generate a new task, complement the event information of the task, and associate the document with the new task, the user's labor is greatly reduced compared to the conventional one, which contributes to the improvement of business efficiency. be able to.

  In addition, the event information is 5W1H + predicate-related information that is related to each other in the document. By taking a machine-readable data format, it is possible to determine the suitability with high accuracy regardless of the machine or application, and versatility. A high-performance document processing method and document processing apparatus can be realized.

  Furthermore, since the suitability of a task is judged based on the event information extracted from the document and the event information of the task, and the task to be predicted is predicted, the existing objective and high accuracy compared to the case where the user performs manually. Association with tasks can be realized.

  In addition, when a new task is generated, information necessary for the new task does not need to be input by the user and can be automatically acquired, so that the labor and cost of the user can be reduced.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIG. The same parts as those in the first embodiment described above are denoted by the same reference numerals, and description thereof is also omitted.

  In the first embodiment, the case where the document processing target is an electronic document has been described. However, in the present embodiment, the case where the newly registered document is a paper document will be described. The present embodiment is an example in which an MFP (Multi Function Peripheral) is applied as a document processing apparatus.

  FIG. 11 is a block diagram showing a schematic configuration of a document processing apparatus 50 according to the second embodiment of the present invention. As shown in FIG. 11, the document processing apparatus 50 includes a scanner 60 that is an image reading unit that optically reads a paper document. Also, as shown in FIG. 11, the document processing device 50 follows the document processing program, so that the event information extraction unit 12, the task prediction unit 13, the task association unit 14, the task generation unit 15, and the task registration unit 16, an information storage unit 17, a reading scanning unit 51 that functions as a reading scanning unit, and a character recognition unit 52 that functions as a character recognition unit.

  Further, the document processing apparatus 1 follows the document processing program to store an image storage unit 61, a text string storage unit 62, an event list 22 for storing event information, and a task for storing tasks in the HDD 104 serving as a storage unit. DB23 is formed.

  Next, a different part from the part demonstrated in 1st Embodiment is demonstrated.

  The reading scanning unit 51 reads a paper document via the scanner 60 and stores the image data of the paper document in the image storage unit 61. The reading scanning unit 51 assigns a document ID to the image data of the read paper document.

  The character recognition unit 52 acquires the text portion in the image data of the paper document by using the OCR or the like for the image data of the paper document read by the reading scanning unit 51 and stores it in the text string storage unit 62. FIG. 12 is an example of text stored in the text string storage unit 62. This process corresponds to step S2 in the event extraction process shown in FIG.

  Then, the event information extraction unit 12 reads the text stored in the text string storage unit 62, extracts all event information that can be acquired in the text, and stores it in the event list 22.

  The subsequent syntactic structure acquisition process (step S3 shown in FIG. 4) and subsequent processes are the same as those in the first embodiment, and thus the description thereof is omitted.

  As described above, according to the present embodiment, when event information is extracted from an accessed paper document, the presence / absence of association with an existing task is predicted, and when it is determined that the association is associated with an existing task, By updating the registered document of the association and existing task, it can be predicted that the document is not related to any existing task, and even in that case, it is very easy without the user reading the contents of the document and without trouble and cost It is possible to generate a new task at the same time, complement the event information of the task, and associate the document with the new task. This greatly reduces the user's labor and contributes to the improvement of business efficiency. Can do.

It is a module block diagram of the document processing apparatus concerning the 1st Embodiment of this invention. It is a block diagram which shows schematic structure of a document processing apparatus. It is a schematic diagram which shows an example of an electronic document. It is a flowchart which shows the flow of the extraction process of event information. It is a schematic diagram which shows the example of syntactic structure extraction from the text of the example document of FIG. It is a schematic diagram which shows an example of the description of a knowledge dictionary. It is a schematic diagram which shows the example of extraction of event information. It is a flowchart which shows the flow of a task prediction, task correlation, and a task new registration process. It is a schematic diagram which shows an example of a task correlation process. It is a schematic diagram which shows an example of a task new registration process. It is a block diagram which shows schematic structure of the document processing apparatus concerning the 2nd Embodiment of this invention. It is a schematic diagram which shows an example of a text.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Document processing apparatus 11 Document access means 12 Event information extraction means 13 Task prediction means 13a Task suitability judgment means 14 Task association means 15 Task generation means 16 Task registration means 23 Task storage means 51 Reading scanning means 52 Character recognition means 60 Image reading Means 104 Storage Unit

Claims (19)

In a document processing apparatus that processes a text document,
Event information extracting means for extracting event information representing an event from the text document;
Task storage means for storing event information of existing tasks;
For the event information extracted by the event information extraction unit, a task prediction unit that predicts an association with an existing task using event information of an existing task stored in the task storage unit;
Associating the text document having the event information predicted to be related to an existing task by the task prediction unit with the corresponding existing task, and updating the task stored in the task storage unit Task association means;
A document processing apparatus comprising: When it is predicted that the task prediction means is not related to an existing task, task generation means for newly generating a task using event information of the text document;
A task registration unit for registering a task newly generated by the task generation unit in the task storage unit together with event information of the text document;
The document processing apparatus according to claim 1, further comprising: The task generation unit generates a new task name and event information of a new task using a phrase of the event information acquired from the text document having the event information predicted to be unrelated to an existing task. ,
The document processing apparatus according to claim 2. The task registration unit sets a new registration flag for the task newly generated by the task generation unit.
The document processing apparatus according to claim 2. The event information extraction means extracts the event information using an integrated structure that clearly shows the relationship of the structure obtained by decomposing the text document from the viewpoint of grammar.
The document processing apparatus according to claim 1, wherein the document processing apparatus includes: The event information extraction means is a machine-readable data format, and extracts 5W1H and predicate-related information that are related to each other in the text document as the event information.
The document processing apparatus according to claim 5. The task prediction means includes task suitability determination means for determining the suitability of a task using event information of an existing task stored in the task storage means for event information of the text document,
The task prediction means predicts an association with an existing task based on the suitability determination by the task suitability determination means;
The document processing apparatus according to any one of claims 1 to 6, wherein the document processing apparatus includes: A document access means for accessing a digitized document;
The event information extraction means reads a text document in the electronic document.
8. A document processing apparatus according to claim 1, wherein Reading scanning means for reading a paper document through the image reading means;
Character recognition means for obtaining a text document in image data of a paper document read by the reading scanning means;
With
The event information extraction means reads a text document in the image data.
8. A document processing apparatus according to claim 1, wherein A program for causing a computer to execute document processing for processing a text document,
An event information extraction function for extracting event information representing an event from the text document;
A task storage function for storing event information of an existing task in a storage unit;
For the event information extracted by the event information extraction function, a task prediction function that predicts an association with an existing task using event information of an existing task stored in the storage unit by the task storage function;
The text document having the event information predicted to be related to an existing task by the task prediction function is associated with the corresponding existing task, and the task stored in the storage unit by the task storage function Task association function to update
That causes the computer to execute the program. When it is predicted that the task prediction function is not related to an existing task, a task generation function for newly generating a task using event information of the text document;
A task registration function for registering a task newly generated by the task generation function in the storage unit together with event information of the text document;
The program according to claim 10, wherein the computer is executed. The task generation function generates a new task name and event information of a new task by using a phrase of the event information acquired from the text document having the event information predicted not to be related to an existing task. ,
12. The program according to claim 11, wherein: The task registration function sets a new registration flag for a task newly generated by the task generation function.
12. The program according to claim 11, wherein: The event information extraction function extracts the event information using an integrated structure that clearly shows the relationship of the structure obtained by decomposing the text document from the viewpoint of grammar.
The program according to any one of claims 10 to 13, characterized in that: The event information extraction function is a machine-readable data format, and extracts 5W1H and predicate-related information that are related to each other in the text document as the event information.
15. The program according to claim 14, wherein: The task prediction function includes a task suitability determination function that determines task suitability using event information of an existing task stored in the storage unit for event information of the text document,
The task prediction function predicts an association with an existing task based on the suitability determination by the task suitability determination function.
The program according to any one of claims 10 to 15, characterized in that: Causing the computer to further execute a document access function for accessing the digitized document;
The event information extraction function reads a text document in the digitized document.
The program according to any one of claims 10 to 16, wherein: A reading scanning function for reading a paper document via an image reading means;
A character recognition function for obtaining a text document in image data of a paper document read by the reading scanning function;
Is further executed by the computer,
The event information extraction function reads a text document in the image data.
The program according to any one of claims 10 to 16, wherein: A document processing method in a document processing apparatus for processing a text document,
An event information extraction step of extracting event information representing an event from the text document;
A task storage step of storing event information of an existing task in a storage unit;
For the event information extracted in the event information extraction step, a task prediction step of predicting an association with an existing task using event information of the existing task stored in the storage unit in the task storage step,
The text document having the event information predicted to be related to the existing task in the task prediction step is associated with the corresponding existing task, and the task stored in the storage unit in the task storage step Task association process to update
A document processing method comprising:

Images