JP2001076004A - Help system, data base construction method and information management/retrieval means - Google Patents

Abstract

PROBLEM TO BE SOLVED: To establish a usage at every execution purpose and to shorten and facilitate setting time by a user by preparing plural databases, setting one database to be that for holding common information, installing another data base for showing optimum information to the user and adding or deleting information. SOLUTION: Not only single data base but a database group corresponding to plural purposes is prepared as a database block 1. A bug report update information data base II and a user data base III can be updated at any time. An information management/retrieval mechanism 2 has a mechanism for adding updated information to a network between pieces of information, which are given an initialization and a mechanism of setting/detecting various parameters following the information. Optimum information, satisfying a retrieval condition given from a user 7 through an interface mechanism 3 based on the parameters, is supplied to the user 7 through a display mechanism 4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an online manual and help system provided with a database, and more particularly to a data providing selection system used for the system and a construction type database corresponding to each user.

[0002]

2. Description of the Related Art A conventional help system shown in FIG. 6 will be described below. The help execution unit 1 derives appropriate information from the help database 2 based on the request items entered by the user into the user interface 8 and presents the information to the display unit 9 via the display control unit 5. If the presentation information satisfies the user's request, the information of the help execution unit 1 is sent to the simulation monitoring unit 6 and the simulation start unit 4, and the transfer information is compared with the simulation start information database 3 and the start information is compared. When the condition of the database 3 is satisfied, an appropriate simulation is executed.

The problem with the conventional help system is that information on simulation execution conditions and the like is static, and it is not possible to deal with exceptions or establish a unique use flow. In addition, there is a problem that the search target is limited only by keyword matching in the search, and it takes time.

[0004]

In the prior art, there is no guarantee that information to be searched for by a user is promptly extracted, and there is no guarantee that a search result satisfies a demand. If you do not have the information you need in your search keywords because you rely on keyword matching, you will not be able to get the information, or you will need to take several steps to get the information together You need to derive the information yourself. This also has the consequence that the user must understand the connection of the information and the range of information that can be searched is limited.

[0005] Further, it is difficult to provide help for a collective flow because setting conditions, post-processing, and use order procedures depending on the use environment and purpose are not fixed for each user, and the explanation for each instruction is ended. It is possible. Therefore, it may take time to understand the flow for each purpose of use and to understand unnecessary commands and the like in order to determine necessary information.

[0006] By providing a parameter for indicating the relationship between individual information, it is easier to search for information, and by constructing a flow information and a user-specific database, the usage method within a user group can be reduced. Aim for unification.
Thus, it is an issue to take over the use result and to easily search for the cause of the error.

[0007]

In order to organize information, a plurality of databases are prepared instead of a single database. One of them is common information, that is, a description of each instruction,
The database is used to hold universal information such as usage conditions, and the user is given only access rights. In addition to this, a database to add tool update information, bug information, additional information, supplementary information, etc., and a database for each user or use group are provided in order to present optimal information that meets the needs of users By doing so, we can advance the information and meet the needs of various users. In other words, this makes it possible to add information that is useful for future use, such as the latest information, special use cases, and precautions in that case, or to delete information that is determined to be unnecessary. In addition, a backup information database is provided to provide insurance for emergencies.

[0008] Information related to the purpose of use is prepared as a search item in order to compensate for the lack of information in the instruction information due to the keyword search. This is because it is considered that it is difficult to determine an executable item from only the instruction usage method and to understand its flow. In order to realize this, it is necessary to link a link between the information. As a method of realizing the information, a parameter indicating the association is provided in the information in the common database, and the corresponding information management means 1 and the information based on the information management means 1 are used. Information retrieval means 1 for extracting information
The set. Similarly, as the construction of the database corresponding to the user environment shown above, the user registration case, the special case correspondence, the bug handling means, etc., and the question and answer record obtained through the communication function are associated with the common database. The information management means 2 to be registered in the user-specific database is created, and the information retrieval means 2 for extracting information based on the information management means 2 is set. As information processing means 1 for organizing display priorities for individual information, networking of related information between the common database and the user-specific database is performed. The importance of the information considering whether it is a response to a special exception is added to each information as a parameter. The following types of parameters are prepared. Indicates the type of information. Instruction specifications, setting information before use, purpose of use information, supplementary information,
Update information, bug information. Indicates the link of related information.
Links to used instructions etc. The priority of the information. Information update date. importance. What expresses universality and specialty.

An interface for presenting search items corresponding to a plurality of purposes is provided to facilitate a search.
Thus, the type and the degree of detail requested by the user are determined, a guideline for the search is set, and the search information is selected.
When producing a result for the requested item presented through the dialogue with the user, the user is provided with a skill recognition means for measuring the skill of the user from the access frequency, the difficulty of the requested item, etc., and is determined to be suitable for the skill. To start with the information. Further, an item to be determined as to whether the database is either the common database or the user-specific database or both depending on the requested content is presented at the time of requesting.
As a result, it is possible to present to the user from more optimal information than the parameters related to the information. An information processing means 2 for reconfiguring the priority of the information formed by the information processing means 1 using the skill level recognition means and the request term recognition means. The help system includes the information management means 1 and 2, the information search means 1 and 2, and the information processing means 2.
Suppose you have

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Communication between a user and a search mechanism is performed via a display. The user inputs search items on the display. The search items refer to the following. Search for instruction usage, explain each option, show executable operation, use instruction for each execution operation, show execution condition, troubleshooting, bug information. The result of the search mechanism, and the result of the help execution mechanism using the search mechanism result are presented to the user via the display.

The search mechanism prepares not only a single database but also a group of databases that can support a plurality of purposes as shown in the figure. The basic setting help database cannot be changed, and the bug report update information database and user database can be updated at any time.

FIG. 1 shows an outline of the system. As shown in the database block 1, a database group corresponding to a plurality of purposes is prepared instead of a single database. Database for updating bug report information in the figure,
The user database can be updated at any time.
In the information management and retrieval mechanism (2), a mechanism for adding updated information to a network between information provided as initial settings, a mechanism for performing setting detection of various parameter values accompanying information, and a mechanism based on these parameters And a mechanism for providing the user with optimal information satisfying the search condition given by the user through the interface mechanism through the display mechanism. Update of information is performed as follows. Information is transferred from the user to the information management and retrieval mechanism 2 via the interface 3. At this time, the type of information is set by the user. This indicates whether the information is a case, a flow of a product usage method, or an exception response. Thereafter, a network is established to show the relationship between the information transferred via the interface 3 and the information in the database block 1. This is performed based on the usage command in the information and the definition of the information type by the user. For example, if the information is a case, the degree of detail is set high, and the information destination address of the related instruction is set as a parameter. In addition, if it indicates a flow or an exception response, an information destination address indicating a condition necessary for step execution,
Set the previous and next step information destination addresses as parameters. The information retrieval method will be described later in detail, but as a basic procedure, a retrieval unit is selected based on the user's request items to the interface 3 and the priority of the information is calculated. Information is presented to the user via the display mechanism 4 in descending order based on the calculated priority order. When the product operation is performed based on the presentation information, the execution control mechanism 5 checks whether all the conditions for product execution are satisfied with the information in the database block 1,
When the conditions are satisfied, the execution operation mechanism 6 is started. This result is presented to the user via the display mechanism 4.

The details of the parameters accompanying the information are shown below.

As an indication of the flow of the flow, an executable operation number indicating the operation purpose and an information number indicating the standard steps before and after are given as parameters. Accompany it. In other words, an executable operation number indicating the purpose and at least three parameter blocks indicating the preceding and succeeding steps are prepared to represent the flow relation, and as a remark, the parameter indicating the number of the correspondence information group at the time of exception and the number of the case information number group are indicated. The purpose of the present invention is to provide a product use procedure in accordance with a user's request by preparing two parameters and linking each information group number in a link format.

Parameters representing the level of detail, importance, and difficulty are given as initial values to individual common information.
Even when a change is made, it is possible to return to the initial value as needed. Next, in order to increase the search speed by sifting the information for keyword matching, parameters are prepared that indicate the type of information item, that is, shared or dedicated to the user, top level or detailed, and so on. For example, when viewing the entire flow, the details are omitted and the description of the detailed settings for each command is excluded from the search items.

The parameter value is changed as follows. Prepare a search engine that has been trained based on the concept of Multiple Vector Quantification. The importance level as information is reset by inputting the number of times of access to the information A, the increase / decrease of related matters, and the skill level of the accessing user for each certain period. When adding a case, a new case is added to the link after confirming the existence of a similar case. As a method of confirmation, it is determined whether or not the command is the same method and whether or not the command is the same flow design based on the used instruction, the method of use, and the order.

When conducting training based on the above-mentioned multiple vector quantification, the inputs are the number of accesses, the skill level of the user, the increase / decrease of related matters, that is, the presence / absence of addition of user-specific information. Repeat the back annotation to approach the importance setting and set the parameters of the inner layer. At this time, the setting that regards the number of accesses as important,
By setting the importance of the related items to be important and performing several types of training, it is possible to prepare a plurality of information presentation methods and select the one that seems to be most suitable for the user.

FIG. 2 shows a basic flow of the search. 1
The request to the search mechanism includes a search item request, a keyword search mechanism, a priority determining mechanism, a display mechanism, and a re-search, which are provided by the user via the interface in FIG. Thereafter, based on the search item request, re-detection of the priority of the information of A and determination of the skill of the user of B are performed in parallel. Details of A and B will be described later. Based on the A information priority re-detection result and the B user skill level detection result,
Search for higher priority items. Here, only items higher than the result of operating the priorities for basic matters, detailed information, and special measures in accordance with the skill level of the user are sent to the C search result display. In the C search result display, the information having the highest priority is presented, and if the user's request is not met, the A information priority is reconfigured again.

As shown in FIG. 3, in the search item request, keyword matching, executable operation, instruction specification details, and case presentation are prepared in order to determine the direction of the search. The search efficiency is improved by changing parameters that are important in each search.

In the keyword matching and instruction specification details, the following parameter operation is performed on the information in which the search keyword is matched by using the keyword search mechanism, and the other information is set to a low priority parameter. The flow-related link information before and after resetting the priority parameter for linked related cases to a higher value lowers the priority for those having a non-single specification purpose.

In the search for each executable operation, the parameter of the executable operation number is focused on, and the priority order is set to zero for those that do not match. Instruction flow,
Priority is set in order of instruction summary, instruction details, and related cases.

As shown in FIG. 4, the skill level of the user is determined based on the number of times of access to the requested item and the level of detail of the requested item. The use group, personal name, first access date, last access date, and number of accesses are recorded as a user database, and these are used as reference material for judging the billing item level of the user. Basically, the number of accesses is used as a material for judging the user's skill level, and the two access dates are used as secondary judgment materials for the material. In other words, the user determines the basic attitude that the user who has a greater number of accesses will be more skilled, and determines what kind of information is necessary for the interval between the two access dates and the interval between the last access date and today. Examples include the following: If the number of accesses is small and the interval between the access recording dates is short, it is determined that the user is a beginner, and the information of the basic level, the group case, and the group flow are the highest priority. If it is judged that the number of times of access is a skilled person, if the interval between the two access days is short, the judgment is changed to that of a beginner.If the interval between the last access date and the day is large, the individual and group case are given top priority. Increase the priority of difficult information. When presenting information, the update date is checked, the case is displayed from the latest one, and the bug report is displayed from the oldest information on the same bug. However, when there is a request from the user, it is before the date and time of the bug report There is an option that can be used to hide things.

After setting the priority level of the information for the request item, the priority level is set up based on the user skill level, and the information is presented to the user based on the priority level. At this time, the information determined to be unnecessary by the user is recorded so that the setting change of the importance for the user presented search request item can be performed later.

See FIG. 5 for the Multiple Vector Quantization. The total value obtained by multiplying the input shown in the figure by each weight according to the instruction of the control unit is output.

Output of the first intermediate layer A0 = Σ (input 0 * Wa1, input 1 * Wa2,..., Input n * WaN) Output of the second intermediate layer B0 = Σ (A0 * Wb1, A1 * Wb
2, ...., An * WbN) Output 0 = Σ (B0 * Wc1, B1 * Wc2, ..., Bn * WcN) During training, the calculated output is compared with the expected value. Feedback is performed to change each weight based on the method in the control unit. The output is detected again from the input and compared with the expected value. The expected value is calculated until the calculated result falls within a certain allowable numerical value. In the future, the output calculated based on the final weight result is determined to be correct.

[0026]

According to the present invention, the following can be achieved by a user database and database construction function which did not exist conventionally. First, by registering a case by a user or a use group, it is possible to present a beginner with a more familiar and easy-to-understand product usage method. Secondly, by constructing a database of the above-mentioned cases and flows, it becomes easy to share and unify the usage methods within the usage group. Further, the use of the user-dedicated database and the sharing of the response at the time of the exception can further facilitate the use. As a result of the operation by the above-mentioned example sharing and unification of usage,
Alternatively, when there is a problem in the previous result, it is considered that it is easy to estimate in which procedure the mistake was made from the malfunction information.

By providing a connection between information in a network format, related information can be derived more easily. Further, by providing parameter values such as importance to the information and grouping the information, it is considered that a user can be presented with appropriate information in a short time based on the information as a search mechanism.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of a system.

FIG. 2 is a basic flowchart of a search.

FIG. 3 is a detailed diagram 1 of a search flow.

FIG. 4 is a detailed diagram 2 in a search flow.

[Figure 5] Flow diagram of the Multiple Vector Quantization

FIG. 6 shows a conventional example.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Database block 2 Information management search mechanism 3 Interface 4 Display mechanism 5 Execution control mechanism 6 Execution operation mechanism

Claims (6)

[Claims] 1. A user-specific information and a usage method by having a plurality of database groups of a user-specific database constructed by the user and a common information beta base that holds invariable information as usage information. And a help system that promptly responds to a user's search request from the plurality of database groups by holding a plurality of search methods. 2. The user-specific beta base, which is set up separately from the common information database that collects invariant information, if necessary, added by the user, flow information for each purpose of use within the user group, It has a function to build special correspondence information in the user-specific database. By establishing a usage method of a product by attaching a parameter indicating a relationship between the information in the common information beta base to the registration information, a usage method in a user group is established or unified. Characteristic database construction means. 3. A parameter group A representing a network indicating a relationship between information in a database (address destination of related information, purpose of use, positioning in use flow, degree of detail,
Parameter group B indicating the importance of information for each piece of information and the relationship between instructions and cases (indicating whether it is a basic usage or special usage, information usage frequency, invariant information or standard exception handling , Update date, importance of information,
Information in the database. An information management search means for calculating a priority of information using the network parameters and the individual information importance parameters. 4. A search is facilitated by having a plurality of search request methods. As the plurality of search request methods, a command usage method request, a use purpose request, a case or exception response method request, an update information request and the network parameters to be regarded as important for each request and the individual information 4. The information management search means according to claim 3, wherein the optimum information according to the user's request is quickly detected by calculating the priority of the information from the importance parameter. 5. A user's skill level is calculated based on a user's access frequency, a non-access period, and a difficulty level of information matching a search request, and the user's skill level is calculated based on the calculated user's skill level. 5. The information management search means according to claim 4, wherein presentation information according to the user's skill level is sorted. 6. The information management and retrieval means according to claim 4 manages the common information database and the user-specific beta base constructed by the database construction means according to claim 3, and presents information according to a user's request. Claim 5
It is characterized by comprising an information display mechanism for presenting the highest priority based on the priority calculated using the information management search means described. 2. The help system according to claim 1, wherein the product operation can be performed when the operation required for the product operation presented to the information display mechanism is completed.