JP4826221B2 - Access analysis system - Google Patents

Description

  The present invention relates to a technique for analyzing a user's work state, and more particularly to a technique for presenting distance variation on a document storage structure between documents accessed by a user as analysis information.

  In the project management, the progress is grasped based on the difference between the schedule prepared in advance and the actual results with respect to the schedule, and the appropriate project is taken as necessary to promote the project smoothly.

Here, as an invention that supports the planning of a new project schedule, information related to the work processes that make up the project and divergence information that shows the difference between the schedule and actual performance of each work process are managed in association with each other, and existing projects are managed. An invention has been proposed that supports schedule planning based on deviation information corresponding to each work process (see Patent Document 1).
However, if there is an existing project that can be used as a reference for planning a new project, an appropriate schedule can be formulated using the deviation information for each work process. Since information was not available, it was necessary to formulate a plan independently.
JP 2002-32225 A.

  In order to promote the project smoothly, it is important for the project manager to accurately grasp whether the work by the worker is proceeding smoothly and to solve it promptly if any problem occurs in the work. However, the problem was how to capture the signal of the problem.

  Here, the worker uses a user terminal device such as a desktop computer or a mobile computer to connect to another computer such as a document management server or a Web site, browse necessary documents, and collect information necessary for the work. . In such a work situation, for example, a document that a worker is browsing is a document that is not related to a document that the worker has viewed in the same job in the past. If the document is not highly related to the browsing document of another worker who is performing, it can be said that some problem has occurred in the work performed by the worker.

  The present invention has been made in view of the above-described conventional circumstances, and by presenting information relating to the relationship between a document accessed by the worker and other documents related to the work content of the worker, The object is to facilitate the determination of whether or not the work is smoothly performed and to provide appropriate support promptly when the work is not smooth.

  The document management structure is a hierarchical structure as illustrated in FIG. 1, and a node and each node under the node are connected to form a plurality of hierarchies, and a document file is stored in the hierarchical structure. Managed in association with one of the nodes.

  In general, in order to facilitate document management and work efficiency, for example, nodes A and B in FIG. Corresponds. On the other hand, distant nodes that require six movements to reach, such as node A and node D in the figure, are associated with documents having a lower degree of association than in the case of node A and node B, respectively. That is, it can be said that the degree of association between documents associated with each node is represented by the distance between the nodes.

  Usually, since each document used in the same work has a high degree of relevance, it is associated with each node having a short distance in the hierarchical structure. You can work well with access (viewing, editing, etc.). However, for example, when there is a shortage of information necessary to proceed with the work, a document search is performed by accessing a node beyond the range that has been accessed so far in order to obtain the information. There is a tendency that the access range is wider than the access range.

  In the present invention, as described above, a plurality of documents (including files of various multimedia data such as images and music) are managed in association with nodes that are separated as the degree of association decreases ( In addition to management by the above hierarchical structure, management by network structure etc. is also included), the distance between nodes represents the degree of association between documents associated with each node. It focuses on the fact that the access range changes according to the work state of the worker, and presents the time-series change of the access range to help grasp the work state of the worker.

  Specifically, the present invention accesses a node in an access analysis system that analyzes access by a user using a node management device that manages a plurality of documents in association with each node according to their relevance. Access log management means for recording and managing user identification information, node identification information and access time information in association with each other, input means for accepting designation of the target user, and access to the node accessed by the designated target user Node specifying means for specifying with reference to the log management means, access range calculating means for calculating distance variation between nodes within a predetermined time width for each specified node based on the node management device, and for each predetermined time width Output means for outputting time-series change information of distance variation of Scan analysis system, a program configuring by executing the system in a computer, a method implemented by the system.

Therefore, a user (for example, a project manager) of this system can grasp a time-series change in distance variation (access range) between nodes accessed by the target user, so that the target user can solve work problems. It is possible to quickly detect that a document search is being performed in order to obtain necessary information, and appropriate support can be provided.
Note that a user group may be designated as a target user, and in this case, the working state of the entire group can be grasped.

The distance variation calculated by this system represents the degree of variation in distance between nodes accessed within a predetermined time width, and each calculated for all combinations of any two nodes in each accessed node. Calculated based on the distance between nodes.
That is, the access range calculation means represents the inter-node distance according to the number of via nodes between the nodes as a representative (a maximum value, an average value, or a representative value obtained by another statistical method).
Further, the distance between nodes may be calculated in consideration of a predetermined weight for the transit node. For example, a weight is added according to the depth of the hierarchy where the transit node is located in the hierarchical structure, By adding weights in accordance with the importance of the nodes, the distance between the nodes more accurately represents the relationship between the corresponding documents.

Here, the access analysis system may output other index values on the same screen in addition to the distance variation.
For example, it further comprises access number acquisition means for acquiring the number of accesses by the target user per predetermined time width with reference to the access log management means, and the output means displays time-series change information on the number of accesses by the target user on the same screen. Output in contrast.
In addition, for example, it further includes a node number acquisition unit that acquires the number of nodes accessed by the target user for each predetermined time width with reference to the access log management unit, and the output unit changes the time series of the number of nodes accessed by the target user. Output information on the same screen.
In this way, since other index values are output in comparison on the same screen, it is possible to compare the time series change of distance variation with the time series change of other index values, and more accurately work state Can be analyzed.

In the access analysis system, the input unit accepts the designation of a plurality of target users, the node specifying unit specifies the node accessed by each target user, and the access range calculation unit specifies a predetermined time for each node specified The distance variation between nodes within the width is calculated for each target user, and the output means outputs the time-series change information of the distance variation calculated for each target user on the same screen. May be.
Thereby, the working state of a certain target user can be grasped based on comparison with other target users. It is also possible to determine whether or not appropriate information sharing is achieved among the target users. For example, it is possible to determine that a target user whose distance variation is small compared to other target users is not sufficiently shared.

  In addition, the output means may output the time-series change information of the ideal distance variation set in advance on the same screen, and may output the work according to the contrast with the ideal distance variation that should originally be. The smoothness can be grasped.

  The node specifying means preferably employs a node that satisfies a predetermined condition. For example, a node with only one access may have been accidentally accessed, so it is possible to perform a more appropriate analysis by excluding such a node from the access node. Since the number of nodes used is reduced, the processing speed can be improved.

Here, for example, when using documents created in past projects, or when working with documents owned by each department in a project consisting of multiple departments, the work is going smoothly. There may be a case where the distance variation becomes large.
Therefore, in such a case, in order to correct the document storage structure and reduce the distance variation, a virtual link attaching means for providing a virtual link for reducing the distance between the nodes managed by the node management apparatus is provided. You may make it prepare.

  According to the present invention, the distance variation between corresponding nodes on the document storage structure relating to each document accessed by the target user is output as time-series change information, so that the user of this system can Based on the change, it can be quickly determined whether or not the work of the target user is smooth, and if it is not smooth, appropriate support can be provided.

The invention will be described in detail based on one embodiment.
FIG. 2 shows an access analysis system according to an example of the present invention.
The access analysis system of this example has a node management table 2 for managing a plurality of documents in association with each node constituting a hierarchical structure as shown in FIG. 1, and the document access status is associated with the document. An access log management table 1 that is managed as an access log to each node, an input unit 3 that receives designation of analysis conditions, an access log acquisition unit 4 that acquires information related to access by the target user from the access log management table 1, and a target user Access range calculation means 8 for calculating the distance variation between nodes within a predetermined time width with respect to the accessed node, based on the node management table 2, and output means for outputting time-series change information of distance variation for each predetermined time width 10. Ideal value that retains ideal distance variation for each predetermined time width Lifting means 11, and a virtual linking means 12 for imparting a virtual link between nodes node management unit 2 manages shorten the distance between nodes.

Note that the access analysis system of this example uses an access analysis apparatus including the respective functional units 1 to 12, but the access analysis system may be configured using a plurality of computers. For example, another server device such as a document management server or a Web server connected to the network has the node management table 2, monitors access to the node managed by the node management table 2, and records and stores it in the access log management table 1 However, it may be used for analysis. Further, for example, another server device connected to the network may include the access log management table 1 and the node management table 2, and the information held in these tables may be acquired via the network and used for analysis.
In this example, each of the functional units 3 to 12 is configured by the computer executing the program according to the present invention, but may be configured as a dedicated circuit function.

As shown in FIG. 4, the node information managed by the node management table 2 corresponds to the “object ID” for identifying the node, the “parent object ID” for identifying the parent node to which the node belongs, and the node. An “object type” indicating the type of attached document (for example, circulation type, editing type, etc.) is recorded.
In this example, the node management table 1 manages documents by associating documents with each node represented in a hierarchical structure. However, a plurality of documents correspond to nodes that are separated as the degree of association decreases. For example, it may be managed by a network structure.

  As shown in FIG. 3, the access log managed by the access log management table 1 includes an “ID” for identifying the access log, “time” when the document is accessed, “user” who performed the access, and type of access (for example, , “Viewing, editing, etc.”, “object ID” for identifying the node associated with the accessed document, and the like are recorded.

The input unit 3 accepts designation of the analysis target user and the analysis period, and in response to the designation, the node specifying unit 5 included in the access log acquisition unit 4 reads “user” from the access log of the access log management table 1. Node is identified by extracting the access log that corresponds to the target user and “time” is within the analysis period, and obtain the “object ID” of the access log for each predetermined time width (for example, one month) I do.
Here, the node specifying means 5 has a node filtering function to be described later, and excludes nodes that satisfy a predetermined condition.

The access range calculation unit 8 includes an inter-node distance calculation unit 9 that calculates a distance between two corresponding nodes in accordance with designation of two different “object IDs”.
When calculating the distance between nodes, the inter-node distance calculation means 9 calculates the path from the node corresponding to one designated “object ID” to the node corresponding to the other “object ID” in the node management table 2. Identify by reference. First, node information of a parent node having “object ID” as “parent object ID” of node information corresponding to one “object ID” is acquired, and “parent object ID” of the node information of the parent node is set to “object” The process of going back to the node is repeated, such as acquiring a parent / parent node with “ID”. Similarly, the other “object ID” is also performed, and a path is identified by identifying a node that is reached from each other “object ID”. Then, as will be described later, the distance is calculated by adding a weight according to each passing node in the path.

  For each “object ID” for each predetermined time width specified by the node specifying means 5, the access range calculating means 8 calculates the distance between nodes of all combinations of arbitrary two “object IDs”. And the representative value (for example, the maximum value) of the distance between these nodes is calculated as the distance variation.

  The output means 10 outputs the time-series change of the distance variation for each predetermined time width calculated by the access range calculation means 8 in a graph format. Note that the output format of the time-series change information is not limited to the graph format as in this example, but may be other formats such as a table format. The output method may be display output on a display or print output on paper using a printer device.

  Therefore, it is possible to grasp the time-series change of distance variation between nodes accessed by the target user, so that the target user who has been accessing a document within a certain range in the hierarchical structure so far has exceeded the range. By accessing this document, you can see that the work is likely not going smoothly, and you will be able to provide support if necessary.

  In this example, the node specifying unit 5 specifies the nodes accessed by the target user for each predetermined time width, and the access range calculating unit 8 calculates the distance variation for each predetermined time width. The node specifying means 5 specifies the time-series nodes accessed by the target user, and the access range calculating means 8 classifies the time-series nodes for each predetermined time width, and calculates the distance variation for each predetermined time width. It may be.

The specification of the analysis period accepted by the input means 3 is not essential. If no specification is made, for example, the entire period may be treated as the analysis period, or, for example, a preset period may be treated as the analysis period. Good.
The predetermined time width uses a value set in advance in the access log acquisition unit 4, but may be designated by the input unit 3.

A flow of processing by the access analysis system of this example will be described.
Here, a case where the target user: U, the analysis period: January to April is designated by the input means 3, and the predetermined time width is set to one month will be described.
First, the distance variation in January is calculated based on the processing flow shown in FIG.

  The access log acquisition unit 4 sets the target user: U, period: January (January 1 to January 31) as the analysis condition (step S1), and the node identification unit 5 corresponds to the analysis condition. A list of access logs is acquired from the access log management table 1 to identify a node (step S2), node information is acquired from the node management table 2 (step S3), and based on the node information, FIG. 6 and FIG. Are used to perform node filtering, which will be described later, to exclude nodes that are not appropriate for use in calculating distance variations (step S4).

The access range calculation unit 8 obtains any two combinations from the identified nodes (step S5), and the inter-node calculation unit 9 calculates the distance between the nodes according to the calculation procedure described later with reference to FIGS. (Step S6), Steps S5 and S6 are repeated until the distance is calculated for all combinations of two nodes (Step S7). When all the distances are calculated, the access range calculation means 8 calculates the distance variation by the calculation procedure described later with reference to FIG. 12 (step S8).
By performing these steps S1 to S8 by switching the period to each month of February, March and April, the distance variation of each month is calculated and output by the output means 10 as time-series change information. To do.

6 and 7 show a processing flow of node filtering according to this example.
According to FIG. 6, any one node is selected from the nodes specified by the node specifying means 5 in step S2 of FIG. 5 (step S11), and the number of access logs corresponding to the analysis condition to the node, that is, The number of accesses is acquired (step S12). Then, the number of accesses is compared with a preset threshold value (lower limit value) (step S13). If it is smaller than the threshold value, the node is excluded (step S14), and if it is larger than the threshold value, it is left as it is (step S15). . These steps S11 to S15 are performed for all the nodes specified in step S2 of FIG.
As a result, for example, a node that is erroneously accessed, such as a node that is accessed only once, can be excluded.
In step S13, an upper limit value may be set as a threshold value for comparing the number of accesses, and the node may be excluded if it is larger than the threshold value, and may be left as it is if it is smaller than the threshold value.
Thereby, for example, a node that is always accessed and has a very large number of accesses such as the top page of the Web system can be excluded.

Further, according to FIG. 7, any one node is selected from the nodes specified by the node specifying means 5 in step S2 of FIG. 5 (step S21), and node information corresponding to the node is acquired from the node management table 2. The node type (“object type” in FIG. 4) is acquired (step S22). Then, this node type is compared with a preset type (step S23), and if it does not match, the node is excluded (step S24), and if it matches, it is left as it is (step S25). These steps S21 to S25 are performed for all the nodes specified in step S2 of FIG.
In step S23, the node type to be excluded may be specified in advance instead of the node type to be left as described above. If they match, the node may be excluded, and if they do not match, they may be left as they are.
Accordingly, for example, a node that is designated as “circulation type” as a node type and needs to be browsed regardless of work can be excluded.

By applying such node filtering, it is possible to exclude nodes that are not appropriate for use in the calculation of distance variation, so that more appropriate distance variation can be calculated and the processing speed can be improved.
It should be noted that the present invention is not limited to the conditions exemplified above, and other conditions such as excluding nodes whose access frequency is less than the prescribed value may be used, or these conditions may be combined.

  FIG. 8 shows a calculation flow of the distance between the nodes, and the distance between the node A and the node B in the node structure as shown in FIG. 9 is illustrated in FIG. A process of calculating with reference to a weight table in which weight distances corresponding to different hierarchies are set will be described. In this example, the node information in the node management table 2 is set with “hierarchy” indicating the depth of the node in the hierarchical structure, and the weight distance can be acquired based on the “hierarchy”. .

  The inter-node distance calculation means 9 acquires a transit node from the node A to the node B with reference to the node management table 2 (step S31). Then, the initial value of the distance is set to 0, node A is set as the target node, and the following processing is performed until the target node becomes node B while replacing the target node according to the movement path.

  First, it is determined whether the target node is the node B (step S32). If the target node is not the node B, the basic distance 1 is added to the distance (step S33). Then, the “hierarchy” is acquired from the node information of the next via node of the target node (step S34), and the weight distance corresponding to the “hierarchy” is determined based on the weight table of FIG. 10A (step S35). ) And added to the distance (step S36). Then, the relay node next to the target node is set as a new target node (step S37), the determination in step S32 is performed again, and the distance calculation processing in steps S33 to S37 is repeated until the target node becomes node B.

Specifically, as shown in FIG. 9, since it is understood that the node A, the node X, the node Y, the node Z, and the node B are routed in this order, the distance calculation process is performed while replacing these nodes as target nodes. I do.
In the first distance calculation process, the node A is the target node, and the “hierarchy” of the next node X is “1”. Therefore, the weight distance is 10 according to the weight table of FIG. A distance 11 is obtained by adding the distance 1 and the weight distance 10. Next, the same processing is performed by replacing the node X with the target node, and the distance 32 is obtained by adding the basic distance 1 and the weight distance 20 to the previously calculated value 11. Next, the same process is performed by replacing the node Y with the target node, and the basic distance 1 and the weight distance 10 are added to the previously calculated value 32 to obtain the distance 43. Next, the same process is performed by replacing the node Z with the target node, and the distance 44 is obtained by adding the basic distance 1 and the weight distance 0 to the previous calculated value 43. Then, when the node B is replaced with the target node, the processing is ended because the end condition (target node = node B) is met, and the distance 44 is calculated as a result.

  In this way, by calculating the distance by adding the weight distance according to the depth of the hierarchy, for example, when managing documents as a node structure according to the organizational structure of the company, it relates to access between documents within the same section The distance and the distance related to the access between documents across sections, even if the number of via nodes is the same, the calculated distance increases across sections, so the distance variation increases, and other sections This makes it easier to determine the unusual work status of accessing a document.

Note that the distance between the nodes may be calculated by other methods, for example, using a weight table in which a weight distance is set for each node as illustrated in FIG. It may be calculated.
Further, for example, without using such a weight table, the distance may be calculated by adding 1 for each via node as shown in FIG. 11, and the distance between node A and node B in this case Becomes 4.

FIG. 12 shows a flow for calculating the distance variation by the access range calculation means 8.
The access range calculation means 8 calculates the distance between the nodes for all combinations of the two nodes in the specified node by the inter-node distance calculation means 9 (step S41), and sets the maximum value of the distance between these nodes in terms of distance. The variation is calculated (step S42).
In addition, it may replace with step S42 and may use another calculation method, for example, you may calculate the average value of each distance and the representative value obtained using another statistical method as distance dispersion | variation.

FIG. 13 shows an example of a graph showing a time-series change in distance variation (access range) output by the output unit 10.
In this example, in addition to the variation in distance (access range), the number of accesses and the number of access nodes are compared and output. However, these are output by the access log acquisition unit 4 according to an instruction from the input unit 3. It is acquired by the access number acquisition means 6 and the node number acquisition means 7 provided. The access number acquiring unit 6 acquires the number of accesses by the target user to each node specified by the node specifying unit 5 based on the access log for each predetermined time width. The access number acquiring unit 6 acquires the number of each node specified by the node specifying unit 5 for each predetermined time width.

As can be seen from FIG. 13A, the number of accesses and the number of access nodes are constant and do not change, but the access range has suddenly increased in March, and a problem has occurred in the work in March.
According to FIG. 13B, the number of access nodes in March decreased compared to February, while the access range increased conversely, indicating that a problem occurred in work in March. .
Thus, by determining in combination with the time series change of other index values, it is possible to more accurately determine whether or not the work by the target user is smooth.

FIG. 14 shows an example in which ideal performance patterns are compared and output along with time-series changes in distance variation (access range). In this example, the ideal value holding means 11 holds ideal values determined for each phase of project planning, design, evaluation, and development, and in response to an instruction from the input means 3, the ideal value holding means 11 The ideal value is acquired and output in comparison with the distance variation (access range).
Thereby, since it can be compared with the ideal value that should be originally, it can be determined whether the project is proceeding smoothly.

The input means 3 of the access analysis system of this example can accept the designation of a plurality of target users. In this case, the node specifying means 5 specifies a node for each target user, and the access range calculating means 9 The distance variation for each target user is calculated, and the output means 10 outputs the distance variation for each target user in comparison with the same screen.
FIG. 15 shows an example in which a time-series change in distance variation (access range) when a plurality of target users A, B, C, and D are designated is shown. According to the figure, it can be seen that the access range in the evaluation and deployment phase is different for user A compared to users B and C, and the work in the evaluation and deployment phase is not smooth. Further, it can be seen that the user D has a narrower access range than the users A, B, and C, and appropriate information sharing is not achieved.

  The access analysis system of this example includes a virtual link attaching unit 12, and a virtual link that reduces the distance between nodes can be given to the node management table 2. Therefore, for example, by using a document created in a past project, when the distance variation becomes large despite normal work behavior, by adding a virtual link so as to reduce the distance to other nodes The document storage structure is corrected, and the distance variation is reduced.

The virtual linking unit 12 of this example performs virtual linking based on a user instruction, and the user outputs the time series change information of distance variation by the output unit 10 as shown in FIG. The virtual link is instructed with reference to the list of distances between nodes.
According to FIG. 17, the distance between nodes AB is 2, the distance between nodes AC is 14, the distance between nodes AD is 14, the distance between nodes AE is 1, the distance between nodes AF is 13, the distance between nodes AG Is 45,..., It can be seen that the node G is separated from other nodes. Then, in response to the user instructing the linking source node G and the linking destination node, the virtual linking means 12 assigns a virtual link as shown by the broken line in FIG. The distance between the nodes A to F is shortened.
It should be noted that a condition for specifying the node from which the virtual link is assigned and a condition for specifying the destination node are set in advance, and the virtual linking means 12 automatically links based on the condition. You may make it attach.

It is a figure explaining the relationship between each document managed by the hierarchical structure. It is a figure which shows the access analysis system which concerns on an example of this invention. It is a figure which shows the holding information of the access log management table which concerns on an example of this invention. It is a figure explaining the holding information of the node management table which concerns on an example of this invention. It is a figure which shows the calculation flow of the distance dispersion | variation which concerns on an example of this invention. It is a figure which shows the processing flow of the node filtering which concerns on an example of this invention. It is a figure which shows the processing flow of the node filtering which concerns on an example of this invention. It is a figure which shows the calculation flow of the distance between the nodes which concerns on an example of this invention. It is a figure explaining calculation of the distance between the nodes concerning an example of the present invention. It is a figure which shows the weight table which concerns on an example of this invention. It is a figure explaining calculation of the distance between the nodes concerning an example of the present invention. It is a figure which shows the calculation flow of the distance dispersion | variation which concerns on an example of this invention. It is a figure which shows the output result which concerns on an example of this invention. It is a figure which shows the output result which concerns on an example of this invention. It is a figure which shows the output result which concerns on an example of this invention. It is a figure explaining the virtual linking of the node which concerns on an example of this invention. It is a figure explaining the virtual linking of the node which concerns on an example of this invention.

Explanation of symbols

1: Access log management table, 2: Node management table,
3: input means, 4: access log acquisition means,
5: Node identification means, 6: Access count acquisition means,
7: Node number acquisition means, 8: Access range calculation means,
9: Inter-node distance calculation means, 10: Output means,
11: Ideal value holding means 12: Virtual link attaching means

Claims (10)

In an access analysis system that analyzes a user's access tendency using a node management device that manages a plurality of documents in association with each node according to their relevance,
An access log management means for recording and managing an access log in which identification information of a user who has accessed a document, identification information of a node corresponding to the document, and access time information are associated with each other;
An accepting means for accepting designation of a target user;
A node identifying unit that identifies a node corresponding to a document accessed by the target user who has received a designation by the receiving unit for each period of a predetermined time width with reference to the access log management unit;
For a combination of two nodes for which a distance is to be calculated, the number of transit nodes related to movement between the nodes related to the combination is obtained with reference to the node management device, and the combination based on the obtained number of transit nodes Having a node-to-node distance calculation function for calculating a distance between nodes representing the degree of association between documents corresponding to the node, and for each period of a predetermined time width, Based on a plurality of inter-node distances calculated by the inter-node distance calculation function for each combination, a maximum value or an average value of these inter-node distances or a representative value obtained by using other statistical methods is used . Access range calculation means for calculating as a value representing the access range by the user;
Output means for outputting time-series change information of the access range by the target user;
An access analysis system characterized by comprising: The access analysis system according to claim 1,
Access for obtaining a number of accesses, which is the number of access logs related to access by the target user for a document corresponding to each identified node for each period of time with reference to the access log management means A number acquisition means;
The output means outputs the time series change information of the number of accesses by the target user in comparison with the same screen. In the access analysis system according to claim 1 or 2,
For each period of a predetermined time width, the apparatus further comprises node number acquisition means for acquiring the number of nodes of each identified node with reference to the access log management means for the period,
The output analysis system, wherein the output means outputs time-series change information of the number of nodes accessed by the target user on the same screen. The access analysis system according to any one of claims 1 to 3,
The input means accepts designation of a plurality of target users,
The node specifying means specifies, for each target user, a node accessed by the target user,
The access range calculation means calculates, for each target user, a value representing an access range by the target user based on each identified node,
The access analysis system characterized in that the output means outputs the time-series change information of the access range for each target user in comparison with the same screen. The access analysis system according to any one of claims 1 to 4,
The access analysis system characterized in that the output means outputs time-series change information of an access range preset as an ideal value on the same screen. The access analysis system according to any one of claims 1 to 5,
The node analysis unit employs a node that satisfies a predetermined condition. The access analysis system according to claim 7,
The inter-node distance calculation function of the access range calculation means calculates the inter-node distance in consideration of a predetermined weight for the transit node. The access analysis system according to any one of claims 1 to 7 ,
Linking at a position where an inter-node distance between the specified node and another node among the specified nodes is shorter than that of the link source node, for the link source node selected from the specified nodes A virtual link attaching means for assigning a virtual link to the previous node;
The access range calculation means uses the inter-node distance between the link destination node and the other node as the inter-node distance between the link source node and the other node. An access analysis system, characterized in that an inter-node distance between a node and the other node is shortened to reduce an access range by the target user. An access analysis system including an access log management unit, a reception unit, a node specification unit, an access range calculation unit, and an output unit manages a plurality of documents in association with each node according to their relevance. In a method of analyzing a user access tendency using a node management device
The access log management means records and manages an access log in which identification information of a user who has accessed a document, identification information of a node corresponding to the document, and access time information are associated with each other;
The accepting means accepts designation of a target user;
The node specifying unit specifies a node corresponding to a document accessed by the target user who has received designation by the receiving unit for each period of a predetermined time width with reference to the access log management unit,
The access range calculation means acquires, for the combination of two nodes that are distance calculation targets, the number of via nodes related to movement between the nodes related to the combination with reference to the node management device, and the acquired via node A node-to-node distance calculation function that calculates an inter-node distance that represents the degree of association between documents corresponding to the nodes related to the combination based on the number, and is specified for the period for each period of a predetermined time width. Based on a plurality of inter-node distances calculated by the inter-node distance calculation function for each combination of two nodes in each node, the maximum or average value of these inter-node distances or other statistical methods are used. calculating a representative value which is a value representing the access range of the user of the object,
The access analysis method, wherein the output means outputs time-series change information of an access range by the target user. A node management device that manages a plurality of documents in association with each node according to their relevance, the identification information of the user who accessed the document, the identification information corresponding to the document, and the access time information In a program configured by causing a computer to execute an apparatus that analyzes an access tendency of a user with reference to an access log management apparatus that records and manages an access log,
A reception function that accepts designation of the target user;
A node specifying function for specifying a node corresponding to a document accessed by the target user who has received a specification by the receiving function for each period of a predetermined time width with reference to the access log management device;
For a combination of two nodes for which a distance is to be calculated, the number of transit nodes related to movement between the nodes related to the combination is obtained with reference to the node management device, and the combination based on the obtained number of transit nodes Having a node-to-node distance calculation function for calculating a distance between nodes representing the degree of association between documents corresponding to the node, and for each period of a predetermined time width, Based on a plurality of inter-node distances calculated by the inter-node distance calculation function for each combination, a maximum value or an average value of these inter-node distances or a representative value obtained by using other statistical methods is used . An access range calculation function for calculating as a value representing the access range by the user;
An output function for outputting time-series change information of the access range by the target user;
An access analysis program characterized in that it is configured by causing a computer to execute.

Images