JP5154109B2 - Document search apparatus, method, and program - Google Patents

Description

  The present invention relates to an information retrieval technique for structured documents.

With the spread of computers and the development of network technology, the exchange of electronic information via networks has become popular. As a result, many of the business processes that have been conventionally performed on a paper basis are being replaced by network-based processes. Particularly, in recent years, many document files have been created as structured documents called XML (eXtensible Markup Language), HTML (Hyper Text Markup Language), and XHTML (eXtensible HyperText Markup Language). Advances in network technology and the spread of structured documents with excellent information searchability have drastically reduced information acquisition costs.
JP 2006-048536 A

Normally, in the document search process, data search conditions are input, and a document including data that meets the search conditions is specified. When a document is specified, the user reads the contents of the document to confirm whether or not the requested information exists. When a plurality of documents are specified, some order is generally given to the plurality of documents, and a structured document is generally displayed according to the ordering. Many users generally read content from documents displayed at the top.
The present inventor pays attention to the user's load accompanying this reading, and in order to further improve the information acquisition efficiency, it is important to identify a document that has a high possibility of containing the requested information with high accuracy and order it at the top. I thought it was.

  The present invention has been completed on the basis of the above-mentioned attention by the present inventor, and its main purpose is information indicating a structured document corresponding to a predetermined search condition (hereinafter referred to as “search”). It is an object of the present invention to provide a technique for rationally ordering structured documents included in search result information so as to enhance user convenience when displaying “result information”.

  One embodiment of the present invention relates to a document search apparatus that searches a database holding a structured document whose data position is specified by a path expression based on a hierarchical structure of tags, and outputs the search results in order. In this apparatus, as a search condition for a structured document, a search character string is received as a static condition receiving unit for detecting a static condition setting input for ordering by a user, a static condition holding unit for holding a static condition, Search request receiving unit that detects input by the user of search request information including dynamic conditions for ordering and search the structured document including the search character string from the database, and obtain the search result as search result information A search execution unit, a dynamic score calculation unit that calculates a degree of matching between a plurality of structured documents included in the search result information and a dynamic condition as a dynamic score, and a plurality of structured documents included in the search result information; A static score calculation unit that calculates a degree of conformance with a static condition as a static score, a document score assignment unit that calculates a document score for each structured document based on the dynamic score and the static score, and a structured document Attached to And an order determination unit that determines the order of the plurality of structured documents included in the search result information based on the document score, and the priority order in the determined order of the plurality of structured documents included in the search result information. And a search result output unit for displaying on the screen.

  According to this configuration, a static condition based on a predetermined evaluation criterion is reflected in a static score, and a dynamic condition based on the subjectivity of a user who executes a search (hereinafter referred to as “search execution user”) is changed. The final document score is calculated by reflecting it on the target score, and the structured documents included in the search result information can be ordered. This makes it easier to rationally order the structured documents included in the search result information so that the search execution user's request is satisfied while being based on a predetermined evaluation criterion.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, a system, a program, a recording medium, etc. are also effective as an aspect of the present invention.

  According to the present invention, when displaying the search result information in the search of the structured document, it becomes easy to reasonably order the structured documents included in the search result information so as to enhance the convenience for the user.

  The document search apparatus according to the present embodiment calculates a dynamic score based on a dynamic condition and a function that calculates a static score based on a static condition for a plurality of structured documents included in search result information. A function for calculating the document score based on the static / dynamic score, and a function for determining the order of the structured document based on the document score. Thereby, the structured documents included in the search result information are rationally ordered, and a document search function that is highly convenient for the search execution user is provided.

  Documents included in the database can be classified on the basis of various evaluation criteria, such as documents for external disclosure and documents for internal disclosure, documents with high reliability, and documents with low reliability. The static condition is a condition indicating the position of a document in the entire document group (hereinafter referred to as “corpus”) included in the database. For example, when a document search is executed from the corpus under the static condition of “document with high reliability”, a static score higher than that of a document currently being reviewed internally is given to a document for external publication. Alternatively, a document created by the president may be given a higher static score than a document created by a general employee. The static score here indicates the degree of fitness of each document with respect to a predetermined static condition. A specific method for calculating the static score will be described later.

  The static condition is specified by a document registration user who is a user who registers various structured documents in the database. The document registration user here includes not only a user who actually registers a structured document but also a user who registers only a static condition.

  On the other hand, the dynamic condition is a condition for measuring the degree of matching between a reference based on the subject of the search execution user and a certain document. For example, in addition to the search character string, when a document search is executed from the corpus under the dynamic condition of “first priority: title area, second priority: keyword area”, a document including the search character string in the title area Only a keyword area is given a higher dynamic score than a document including a search character string. Alternatively, a dynamic score may not be given to a document that includes a search character string only in the body area. The dynamic score here indicates the degree of suitability of each document for the dynamic condition specified by the search execution user.

  Note that the number of appearances of the search character string in the document may be specified as the dynamic condition. In this case, a dynamic score is given based on the number of appearances of the search character string. Other specific methods for calculating the dynamic score will be described later.

FIG. 1 is a schematic diagram showing an outline of processing of the document search apparatus 100.
When the search execution user inputs search request information including a search character string and dynamic conditions to the document search apparatus 100, the document search apparatus 100 searches the database 300 for a structured document including the search character string, and The search result information including the structured document is acquired. The search character string is a character string having a certain meaning, and may be a natural sentence or a keyword. The structured document in the database 300 is a document structured by tags such as an XML document or an XHTML document.

  The search ranking control unit 140 of the document search apparatus 100 calculates a dynamic score / static score for each structured document included in the search result information. Here, the dynamic score calculation is performed based on the dynamic condition, and the static score calculation is performed based on the static condition of the static condition holding unit 120. Then, a document score is calculated from the dynamic score / static score, and each structured document is ordered based on the document score. The document score here is a score that is calculated by a predetermined evaluation function based on the dynamic score / static score of each structured document and serves as a reference for ordering. Finally, the document search apparatus 100 returns a plurality of structured documents with priorities as search result information to the search execution user.

FIG. 2 is a functional block diagram showing the configuration of the document search apparatus 100. As shown in FIG.
Each block shown here can be realized in hardware by an element such as a CPU of a computer or a mechanical device, and in software it is realized by a computer program or the like. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by a combination of hardware and software.

The document search apparatus 100 includes a user interface processing unit 110, a static condition holding unit 120, a search execution unit 130, and a search ranking control unit 140.
As described above, the database 300 holds a structured document to be searched and is connected to the document search apparatus 100 via the communication network 200. Here, the communication network 200 includes various communication means such as the Internet, LAN, and WAN. Furthermore, the document search apparatus 100 and the database 300 may be realized on the same hardware.
The search execution unit 130 searches the database 300 for a structured document including the search character string included in the search request information from the user, and acquires the search result as search result information.

  The user interface processing unit 110 is in charge of processing related to the entire user interface, such as input processing from the user and information display to the user. In the present embodiment, description will be made assuming that the user interface service of the document search apparatus 100 is provided by the user interface processing unit 110. As another example, the user may operate the document search apparatus 100 via the Internet. Alternatively, the client application may perform input / output with the document search apparatus 100 via a communication network. In either case, a communication unit (not shown) receives search request information from the user terminal or client application, and transmits search result information executed based on the request to the user terminal or client application.

The user interface processing unit 110 includes a static condition receiving unit 112, a search request receiving unit 114, and a search result output unit 116.
The static condition receiving unit 112 receives a static condition input operation from a document registration user, and stores the static condition in the static condition holding unit 120 described above. The search request receiving unit 114 receives an input operation for search request information from a search execution user. The search result output unit 116 prioritizes a plurality of structured documents included in the search result information in the order determined by the search ranking control unit 140 and displays the structured documents on the screen of the search execution user.

  The “search result information” displayed on the screen of the search execution user in the present embodiment is provided as a page indicating a predetermined number of document names and a part of the document from the top of the document score. Other display modes may be used. For example, a document with a higher document score may be displayed in a font larger than other documents, or displayed in a more prominent place compared to other documents. Further, only the document name may be provided for some documents according to the document score.

The search ranking control unit 140 whose function has been described with reference to FIG. 1 includes a dynamic score calculation unit 142, a static score calculation unit 144, a document score assignment unit 146, and an order determination unit 148.
The dynamic score calculation unit 142 calculates the degree of matching between the plurality of structured documents included in the search result information and the dynamic condition as a dynamic score. The static score calculation unit 144 calculates the degree of matching between the plurality of structured documents included in the search result information and the static conditions as a static score. An example of calculation of the dynamic score and calculation of the static score will be described later.
The order determining unit 148 determines the order in which a plurality of structured document names included in the search result information are arranged based on the document score of the structured document calculated by the document score assigning unit 146. The order determination method is not limited to some algorithms. For example, the order may be determined in descending / ascending order of the document score, or the order may be determined by adjusting the document score and other parameters.

  The document score assigning unit 146 calculates a document score for each structured document based on the dynamic score / static score for each structured document. The document score may be calculated in any way as long as the order of each structured document is output in a format that can be determined by the document score. For example, it may be a simple addition of a dynamic score or a static score, or may be calculated using integration or various known arithmetic / statistical methods.

  Furthermore, a weighting setting indicating the ratio of each of the dynamic score and the static score contributing to the document score is given as a part of the dynamic condition, and the document score assigning unit 146 follows the dynamic setting and the static score according to the weighting setting. The ratio at which the target score contributes to the calculation of the document score may be changed. As a result, the weighting of the dynamic score based on the subjective criteria of the search execution user and the static score based on the objective criteria can be adjusted, and the optimal ordering according to the status of the search execution user and common ordering policies, etc. Can be realized.

  An example of document score calculation based on the weight setting is shown below. For example, when the weighting setting is dynamic: static = 1: 1, the dynamic score is 20 points out of 50, the static score is 30 points out of 50, and the document score is 100 points. In some cases, since there is no difference in the degree of contribution to the document score, it is added as it is, and the document score becomes 50 points. As another example, if the score is left as it is and only the weighting setting is dynamic: static = 2: 3, the contribution degree of the dynamic score is 40%. The minute is 16 points. Similarly, since the contribution of the static score is 60%, the contribution of the static score is 36 points when normalized to a full score of 60. As a result, the document score is 52 points.

  As described above, if the ordering on the subjective basis of the search execution user is to be increased, the contribution of the dynamic score to the document score may be increased. Conversely, if the ordering based on objective criteria such as reliability and importance indicating the positioning of the structured document in the corpus is to be increased, the contribution of the static score to the document score may be increased.

As an extreme example of the weighting setting, a case where the static score weighting is set to 100% and the dynamic score weighting is set to zero will be described. In this case, the dynamic condition included in the search request information loses meaning, and only the structured document is narrowed down based on the search character string. The document score assigning unit 146 invalidates the contribution of the dynamic score to the document score, and calculates the document score based only on the static score. Under this weighting setting, since the dynamic score of the structured document is not reflected in the document score, a structured document with a large static score becomes a structured document with a large document score as it is. This setting is useful when you want to obtain search result information in which structured documents are ordered based only on objective criteria such as reliability and importance.
As an extreme example, the case where the dynamic score weighting is set to 100% and the static score weighting is set to zero will be described. In this case, the document score assigning unit 146 invalidates the contribution of the static score to the document score, and calculates the document score based only on the dynamic score. This setting is useful when it is desired to obtain search result information in which structured documents are ordered based only on the subjective criteria of the search execution user.

FIG. 3 is a flowchart showing the flow of processing in the document search apparatus 100. This figure shows the flow of processing by each functional block described so far. Note that the static condition is input by the user when the structured document is registered, and is held in the static condition holding unit 120 via the static condition receiving unit 112.
The search request receiving unit 114 detects input of search request information by the user (step S1). The search request information includes a search character string and a dynamic condition. The search execution unit 130 searches the database 300 for a structured document including the search character string, and acquires the search result as search result information (step S2). The dynamic score calculation unit 142 calculates a dynamic score from each structured document and dynamic conditions included in the search result information (step S3). However, if the weighting of the dynamic score is zero, the process of S3 is skipped. The static score calculation unit 144 calculates a static score from each structured document and static conditions included in the search result information (step S4). However, when the static score weight is zero, the process of S4 is skipped. The document score assigning unit 146 calculates the document score of the document from the dynamic score / static score of each structured document (step S5). The order determination unit 148 determines the order of the plurality of structured documents included in the search result information from the document score of the structured document (step S6). Based on the determined order, the search result output unit 116 prioritizes each structured document and displays it on the screen (step S7).

  Hereinafter, the calculation method of the dynamic score and the static score will be described with specific examples. In the following, an XML document is shown as an example of a structured document and XPath is shown as an example of a path expression.

In this specification, a syntax that indicates a tag path based on a tag hierarchical structure, such as an XPath expression, is referred to as a “path expression”. For example, assume that there is an XML document as follows.
<doc>
<title></title>
<gaiyou>
<comment> Comment 1 </ comment>
</ gaiyou>
<body>
<chapter1>
<comment> Comment 2 </ comment>
</ chapter1>
</ body>
</ doc>
Here, the “comment” tag related to “comment 1” can be specified by the path expression “/ doc / gaiyou / comment”, and the “comment” tag related to “comment 2” is “/ doc / body / chapter1 / comment”. Can be specified by the path expression

  XPath is a notation that can handle ellipsis. For example, the path expression “// comment” means a “comment” tag somewhere in the XML document, and two “comment” tags are specified in the XML document. The path expression “/ * / * / comment” means a “comment” tag somewhere in the third level of the XML document. In the XML document, the “comment” tag related to “comment 1” is Will be specified.

If the user can describe a path expression without an ellipsis, a specific tag of the structured document can be uniquely specified. However, it is not always possible to know the path formula accurately. For example, even if you know that the tag you want to specify is the "comment" tag under the "body" tag, what kind of tag is between the "body" tag and the "comment" tag I don't understand. In such a case, a path expression including an ellipsis as described above that allows flexible tag specification using the tag name as a key is useful. A path expression that includes an ellipsis is also useful when you want to specify multiple tags at once using the tag name as a key.
In this specification, a path expression that includes an ellipsis is referred to as an “incomplete path expression”, and a path expression that does not include an ellipsis is referred to as a “complete path expression”. A complete path expression is a path expression that can uniquely identify a specific tag in a structured document, and an incomplete path expression is insufficient to uniquely identify a specific tag in a structured document. You can say that. In other words, the incomplete path expression can be said to be a flexible path expression that can specify a plurality of tags at once.
The “path expression” in this specification is a concept including both a complete path expression and an incomplete path expression.

Example 1
The dynamic conditions of this embodiment include a path expression that designates a specific area in the structured document. Hereinafter, a path expression designated as a dynamic condition or a static condition will be specifically referred to as a “designated path expression”. The search execution unit 130 searches the database 300 for a structured document including a search character string in a specific area (hereinafter referred to as “specified path area”) specified by the specified path expression, and searches the search result. Obtained as result information.

  4 and 5 are examples of structured documents registered in the database 300. FIG. For example, it is assumed that the search execution user designates “extension number” and a designated route expression “/ doc / keywordt” as a search character string. In this case, since the structured document of FIG. 4 includes the search character string “extension number” in the designated path area, the structured document is included in the search result information. On the other hand, the structured document of FIG. 5 does not include the search character string in the designated route area, and thus is not included in the search result information. That is, only the structured document including the search character string in the designated route area is included in the search result information.

According to the present embodiment, the structured information included in the search result information is provided on the condition that the search character string designated by the search execution user is included in a specific area from among a large number of structured documents registered in the database. You can narrow down documents. As a result, the convenience when the user reads the search result information is improved.
For example, if the search execution user designates the search character string as “extension number” and does not specify the search range, the structured document in FIG. 5 also includes the search character string in “/ doc / body” and is included in the search result information. Will be. However, by providing the designated path expression “/ doc / keywordt”, the structured document of FIG. 5 that does not include the search character string “extension number” in the designated path area is not included in the search result information. If this search execution user needs an extension number table, the method according to this embodiment reduces the load of reading the search result information by eliminating the document according to FIG. 5 unnecessary for the search execution user. Can be reduced.

As a modification, the search execution unit 130 may search the structured document including the search character string from the database 300 and acquire the search result as search result information. At this time, the dynamic score calculation unit 142 adjusts the dynamic score with a predetermined score adjustment value only for the structured document including the search character string in the designated path area.
In the above example, both the structured document in FIG. 4 and the structured document in FIG. 5 include the search character string “extension number” in the document, and thus are included in the search result information. Since the structured document in FIG. 4 includes “extension number” in the designated route area, the dynamic score calculation unit 142 increases the dynamic score of the structured document in FIG. 4 by adding a predetermined score adjustment value or the like. Let On the other hand, since the structured document of FIG. 5 does not include the “extension number” in the designated route area, the dynamic score calculation unit 142 does not increase the dynamic score of the structured document of FIG. As a result, the structured document of FIG. 4 is more likely to be ordered higher. However, since the structured document in FIG. 5 is also included in the search result information, the search execution user can check the structured document in FIG. 5 from the display of the search result information.

(Example 2)
The dynamic conditions of the present embodiment include a plurality of types of designated route expressions and score adjustment values for each designated route expression. The dynamic score calculation unit 142 adjusts the dynamic score of the structured document including the search character string in the designated route area by the score adjustment value corresponding to the designated route expression.

The search execution user specifies "extension number" as the search character string, and (specified route expression, score adjustment value) = (/ doc / title, 10 times), (/ doc / body, 1.5 times) as the dynamic condition Suppose that
For example, when a certain structured document includes a search character string in the designated route area, one point is added as a dynamic score. If “extension number” is included in “/ doc / title”, 10 times that number is added. On the other hand, if “extension number” is included in “/ doc / body”, 1.5 points are added. If “/ doc / body” includes “extension number” three times, 1.5 × 3 = 4.5 points are added. That is, the dynamic score tends to be higher when “/ doc / title” includes “extension number” than “/ doc / body” includes a plurality of search character strings “extension number”. In other words, the search execution user thinks that a structured document including “extension number” in “/ doc / title” is highly likely to be a desired structured document. According to the present embodiment, it is possible to execute a document search in a form that closely reflects the search policy of such a search execution user.

  In this embodiment, the integration is shown as an example of the score adjustment of the dynamic score. However, this may be addition, or the dynamic score of another document may be subtracted. Furthermore, the score may be adjusted by other known arithmetic / statistical score calculation methods. In the case of addition, for example, the score addition value corresponding to the designated path expression is added to the dynamic score of the structured document. In the case of subtraction, for example, the score subtraction value corresponding to the designated path expression is subtracted from the dynamic scores of other structured documents. The same applies to other embodiments.

As described above, an incomplete path expression can be specified as the specified path expression. When an incomplete route expression is designated as a dynamic condition, there may be a plurality of designated route areas. In this embodiment, the presence or absence of a search character string is determined for each designated route area, and the dynamic score is adjusted. For example, it is assumed that the following XML document is included in the search result information.
<doc>
<title></title>
<gaiyou>
Please correct <comment> TODO ○○. </ comment>
</ gaiyou>
<body>
<chapter1>
<comment> TODO Please delete △△. </ comment>
</ chapter1>
</ body>
</ doc>
Here, when “// comment” is designated as the designated route expression, the designated route area becomes a data area of “/ doc / gaiyou / comment” and a data area of “/ doc / body / chapter1 / comment”. If the search character string is “TODO”, since the search character string is included in each designated route area, the dynamic score of the XML document is adjusted twice.
As shown here, the user can flexibly specify a plurality of areas in the structured document using the tag name as a key by specifying the incomplete path expression as the specified path expression. The same applies to other embodiments. In this embodiment, the dynamic score can be adjusted based on the value of each designated area.

(Example 3)
The dynamic condition of the present embodiment includes dynamic score area designation information for designating a dynamic score area as a partial area in the structured document. The dynamic score area designation information may be information for designating an area indicated by a node located at the end in the tag structure of the structured document, or by a predetermined node of the structured document and a node near the predetermined node. It may be information for designating an area to be displayed, or information for designating an area to be shown as an entire structured document. The dynamic score calculation unit 142 calculates a dynamic score for the specified dynamic score area.

FIG. 6 is a diagram for explaining an area designated by the dynamic score area designation information according to the present embodiment. The search execution user can specify a region to which a dynamic score is to be assigned as follows.
1) Region 412 indicated by the node located at the end in the tag structure of the structured document
2) Area 410 indicated by a predetermined node of the structured document and a node near the predetermined node
3) Area 400 shown as a whole of the structured document
In addition, the area indicated by the predetermined node of the structured document and the node in the vicinity of the predetermined node in 2 above may be specified by the search execution user, or the document search apparatus 100 automatically calculates the adjacent node. You may go.

  A neighboring node refers to a node that is close to a specific node in a structured document in a hierarchical structure. For example, in the structured document of FIG. 4, when “/ doc / title” is designated by the search execution user, “/ doc / author” which is the adjacent node having the same hierarchy depth in the structured document. Also, “/ doc / keywordt” may be determined as a neighboring node.

  A specific example of this embodiment will be shown. In the structured document of FIG. 4, when the search character string is “in-house document” and the dynamic score area designation information is “/ doc / title” node, the dynamic score calculation unit 142 displays the structured document. Do not increase the dynamic score. However, if the dynamic score area specification information is the “/ doc / title” node and its neighboring nodes, and “/ doc / keywordt” is the neighboring node of “/ doc / title”, The static score calculation unit 142 increases the dynamic score of the structured document.

  According to the present embodiment, when the search execution user designates an area for calculating the dynamic score, the structured document included in the search result information is easily displayed in an order reflecting the intention of the search execution user. For example, if a specific node is known and it is desired to order by the number of appearances of the search character string in the node, the above 1 may be specified. If it is desired to order by the number of appearances of the search character string in the entire document, the above-mentioned 3 may be specified. Furthermore, if you think that the approximate position of the search string is known but may be different, order by the number of occurrences of the search string in the partial area that also includes neighboring nodes that are close in the hierarchical structure. If you want to do this, specify 2 above. Thus, it is useful in that the ordering can be flexibly changed in accordance with the understanding of the structure of the structured document of the search execution user and the need for ordering. Further, the document search apparatus 100 automatically calculates the neighboring nodes, so that the understanding of the structure of the search execution user's structured document can be supplemented as in the above-described example, and the convenience of the search execution user can be further enhanced.

Regarding the calculation of neighboring nodes, a node located at a fixed value distance from a predetermined node may be fixedly determined as a neighboring node. Alternatively, a predetermined node may be used as a reference node, a comparison node that is a candidate for a neighboring node may be selected from a structured document, and a neighboring node may be determined based on the proximity of the reference node and the comparison node in the hierarchical structure Good.
The latter example will be described below. Here, the proximity of the position of the reference node and the comparison node on the hierarchical structure is calculated as the tag adjacency, and the comparison node whose tag adjacency is equal to or greater than a predetermined threshold is determined as the neighborhood node. In calculating the tag adjacency, first of all, the tag at the deepest level when viewed from the root node is identified as the common tag among the parent tags common to the reference tag and the comparison tag. Next, the depth element value is calculated based on the depth of the hierarchy of the common tag, the reference tag, and the comparison tag. Then, the order element value is calculated based on the path from the common tag to the reference tag and the path from the common tag to the comparison tag. Finally, the tag adjacency is calculated by a weighted average of the depth element value and the order element value.

The depth element value increases as the common tag is located deeper and the difference between the common tag depth and the reference tag depth, and the difference between the common tag depth and the comparison tag depth are smaller. That is, in the tag hierarchy, the depth element value of the reference tag and the comparison tag, which are closely related to the depth at a deep position, has a characteristic of increasing.
The order element value increases as the common tag is located deeper and the path from the common tag to the reference tag and the path from the common tag to the comparison tag are closer to each other. That is, the order element value has a characteristic that it becomes a large value for the reference tag and the comparison tag that are closely related to the route at a deep position in the tag hierarchy.
The tag adjacency has a feature that it increases as the depth element value increases, and similarly increases as the order element value increases.

  Usually, the tag hierarchy often defines the sentence structure as it is, and the contents of the document are structured to some extent by the tag hierarchy. For example, the deeper the common tag, the more frequently the area indicated in the scope of the common tag is detailed and embodied. Also, the closer the reference tag and comparison tag are to the common tag in terms of depth and path, the more information included in the scope of the common tag and the scope of the comparison tag will be included in the information included in the scope of the common tag. Often, the information in is closely related. The calculation of the neighboring node described above is based on such knowledge, and the range of the neighboring node is rationally specified as compared to the case where the node having a fixed value distance from the reference node is fixed as the neighboring node. It becomes easy.

Example 4
In this embodiment, the static condition includes a path expression that specifies a specific area in the structured document and a score adjustment for adjusting the static score of the structured document that includes the search character string in the specified path area. Contains the value. The static score calculation unit 144 adjusts the static score of the structured document including the search character string in the designated route area by the score adjustment value.

  While holding the static condition, the static condition holding unit 120 creates and holds a set of the document ID of the document including the designated path area, the node ID of the designated path area, and the score adjustment value as the first index. To do. The static score calculation unit 144 acquires, from each structured document included in the search result information, a set of the document ID of the document including the search character string and the node ID of the node including the search character string as a second index To do. The static score calculation unit 144 further compares the document ID / node ID of the first index and the second index, and if they match, the static score calculation unit 144 is structured by the score addition value corresponding to the document ID / node ID. Adjust the static score of the document.

  Assume that the document registration user of the structured document in FIG. 4 specifies the specified path expression “/ doc / keywordt” as a static condition and “50” as the corresponding score adjustment value. If the search execution user designates “extension number” as a search character string, this structured document includes “extension number” in “/ doc / keywordt”. The score adjustment value “50” is added to the static score.

  In this way, by performing score adjustment based on the presence / absence of keywords in the designated route region in the structured document, it is possible to display search result information ordered based on objective criteria. In other words, it is possible to set a common standard in which a keyword related to a structured document to be registered is described in a specific area and ordered according to the matching with the keyword. For example, it is useful when you want to order structured documents included in search result information based on a common policy, such as “Highly order documents that match the search string and keyword tag string”. .

(Example 5)
In the present embodiment, the static condition includes static ordering information indicating a path expression designating a specific node in the structured document and a rule for ordering each structured document based on the value of the specific node. Is included. The static score calculation unit 144 acquires a value of a specific node designated by a path expression from the structured document, refers to the static ordering information, and statically calculates the static value of the structured document based on the value of the specific node. Calculate the score.

The static ordering information in this embodiment indicates a rule for ordering a specific data string such as “ascending order” or “descending order”.
While holding the static condition, the static condition holding unit 120 includes the document ID of the document including the specific node specified by the path expression, the node ID of the specific node specified by the path expression, and the static ordering. A set of information is created and held as a first index. The static score calculation unit 144 acquires the document ID of the document including the search character string from each structured document included in the search result information, and creates it as a second index. The static score calculation unit 144 further compares the document IDs of the first index and the second index, and if they match, the value of the node designated as the node ID held in the first index To get. Then, the static score of the structured document is adjusted with a predetermined adjustment score so as to be ordered based on the value of the node and the static ordering information.

As a static condition, it is assumed that “/ doc / doc_rank” is specified in the path expression and “ascending order” is specified as the corresponding static ordering information. The static score calculation unit 144 acquires “/ doc / doc_rank” of each structured document included in the search result information, and adjusts the static score of each structured document so that it is ordered in descending order. Here, it is assumed that the structured document of FIGS. 4 and 5 is associated with this static condition.
In this case, “/ doc / doc_rank” of the structured document of FIG. 4 is “80”, and “/ doc / doc_rank” of the structured document of FIG. 5 is “200”. Since the static ordering information is “ascending order”, the static score calculation unit 144 adjusts the static score so that the structured document in FIG. 4 is ordered higher than the structured document in FIG. 5. For example, a larger score adjustment value than the static score of the structured document of FIG. 5 is added to the static score of the structured document of FIG.

  In this way, it is possible to display search result information ordered based on objective criteria by performing score adjustment based on the value of a specific node specified by the document registration user in the structured document and the ordering information. Become. In other words, it is possible to set a common standard in which a value for ordering structured documents to be registered is described in a specific node and ordered based on the value of the node and ordering information. For example, it is useful when it is desired to order structured documents included in search result information based on a common in-house policy.

Further in the above example, the value of “/ doc / doc_rank” is set to “20” if the document creation organization is the president's office, “80” if the document creation organization is the general affairs department, If the personnel department, “60”. The static ordering information is assumed to be “ascending order”. In this case, the static score is adjusted in the order of the size of “/ doc / doc_rank”, and even if there are many registered documents in the database 300, the document creation organization is “President's office”, “HR department”, “General affairs” Documents are easily ordered in the order of “part”.
As another example, ordering based on a node describing “update date and time”, ordering based on a node describing “walking time from a station” in real estate information, and the like are possible.

(Example 6)
In the present embodiment, the static condition includes a path expression that specifies a specific node in the structured document and information that specifies a value range of the specific node. The static score calculation unit 144 The value of a specific node specified by a path expression is obtained from the structured document, and the static score of the structured document is adjusted based on the information specifying the value and range of the specific node

The information for designating a range in this embodiment is range designating information for designating a range of a specific value such as “above threshold value” or “below the first threshold value and below the second threshold value”.
While holding the static condition, the static condition holding unit 120 includes a document ID of a document including a specific node specified by the path expression, a node ID of the specific node specified by the path expression, and range specification information Is created and held as a first index. The static score calculation unit 144 acquires the document ID of the document including the search character string from each structured document included in the search result information, and creates it as a second index. The static score calculation unit 144 further compares the document IDs of the first index and the second index, and if they match, the value of the node designated as the node ID held in the first index To get. Then, it is determined whether the value of the node is included in the range specified by the range specification information. If included, the static score of the structured document is adjusted with a predetermined adjustment score.

As a static condition, it is assumed that “/ doc / date” is specified in the path expression and “after April 1, 2006” is specified as the corresponding range specification information. In other words, there is a common standard in which new documents after April 1, 2006 are given preferential treatment in terms of static scores and are highly ordered. The static score calculation unit 144 acquires the value of “/ doc / date” of each structured document and compares the value with the range designation information. When the value of “/ doc / date” of the structured document satisfies the condition of the range designation information, the static score calculation unit 144 adjusts the static score of the structured document.
In this case, “/ doc / date” of the structured document of FIG. 4 is “2007/02/01”, and “/ doc / date” of the structured document of FIG. 5 is “2006/01/08”. It has become. Therefore, the static score calculation unit 144 increases the static score of the structured document in FIG. 4 that satisfies the condition of the range designation information by adding a predetermined score adjustment value or the like.

  Regarding the adjustment of the static score, the static score calculation unit 144 determines whether the value of the node is included in the range specified by the range specification information. If included, the static score of the structured document is determined. You may adjust with a different adjustment score according to the value of the node. For example, when the range designation information is greater than or equal to the first threshold and less than or equal to the second threshold, if the value of the node is the same value as the first threshold, the minimum adjustment score is added, and the second The maximum adjustment score may be added when the value is the same as the threshold value, and the weighted average value of the minimum adjustment score and the maximum adjustment score may be added according to the value of the node when the value is between them. That is, a smaller adjustment score may be added as the node value is closer to the first threshold value, and a larger adjustment score may be added as the node value is closer to the second threshold value.

  Furthermore, regarding the adjustment of the static score, the static score calculation unit 144 determines whether or not the value of the node is included in the range specified by the range specification information. The score may be adjusted with a different adjustment score depending on the value of the node. For example, when the range designation information is greater than or equal to the first threshold and less than or equal to the second threshold, if the value of the node is less than the first threshold, the minimum adjustment score is always added, and the second When it is larger than the threshold value, the maximum adjustment score may always be added.

In this way, it is possible to display search result information ordered based on objective criteria by adjusting scores based on specific node values and range specification information specified by document registration users in structured documents. It becomes. In other words, it is possible to set a common standard in which information for ordering structured documents to be registered is described in a specific node, and when the information is included in a specific range, a high order is set. Become. In the above-described example, there is an effect that it becomes easy to rank a new extension number table after April 1, 2006 as a higher rank. On the contrary, it becomes easy to order the old extension number table which is likely not to correspond to the current situation in the lower order.
As another example, it is possible to extract a monthly sales report whose sales performance does not meet the target, or search by specifying a range of walking time from the station in the real estate information.

(Example 7)
In this embodiment, the static condition is associated with a path expression that specifies a specific node in the structured document, a predetermined fixed value to be compared with the value of the specific node, and a predetermined fixed value. The score adjustment value is included. The static score calculation unit 144 acquires a value of a specific node specified by a path expression from the structured document, and when the value of the specific node matches a fixed value, the static score of the structured document is fixed. Adjust by the score adjustment value corresponding to the value.

  When the static condition holding unit 120 holds the static condition, the static condition holding unit 120 sets a set of a document ID of a document including a specific node specified by the path expression and a node ID of the specific node specified by the path expression. Create as 1 index. Then, a fixed value table having a record of a set of the node ID, the fixed value to be compared, and the score adjustment value is created. The static score calculation unit 144 acquires the document ID of the document including the search character string from the structured document included in the search result information, and creates it as a second index. The static score calculation unit 144 further compares the document IDs of the first index and the second index, and if they match, the value of the node designated as the node ID held in the first index To get. Next, it is determined whether the value of the node matches one of the fixed values held in the fixed value table. If they match, the static score of the structured document is adjusted by the score adjustment value corresponding to the fixed value. To do.

As a static condition, it is assumed that “/ doc / author” is specified in the path expression, and the fixed value and the corresponding score adjustment value are specified as follows.
(Fixed value, score adjustment value) = (President A, 100), (B department head, 80), (C section head, 60).
FIG. 7 is an example of a structured document in this embodiment. If the search execution user designates “year-old business guidelines” as a search character string, the search execution unit 130 acquires the structured document of FIG. 7 as search result information. The static score calculation unit 144 acquires the value of “/ doc / author” of each structured document and determines whether the value exists in the fixed value table. In this case, since “B section manager” exists in the fixed value table, the static score of the structured document in FIG. 7 is increased by adding the corresponding score adjustment value 80 or the like.

  In this way, it is possible to display search result information ordered based on objective criteria by adjusting scores based on specific node values and range specification information specified by document registration users in structured documents. It becomes. In other words, it is possible to set a common criterion that, when the value of a specific node in the structured document is a predetermined fixed value, a high ordering is performed. In the above-described example, even when there are a large number of registered documents in the database 300, the value of “/ doc / author”, that is, the creator, is easily ordered in the order of “A president”, “B department manager”, and “C section manager”.

(Example 8)
In the present embodiment, the static condition includes a path expression that specifies a specific node in the structured document and a score calculation expression that calculates a score adjustment value based on the value of the specific node. The static score calculation unit 144 acquires a value of a specific node specified by a path expression from the structured document, calculates a score adjustment value from the value of the specific node and the score calculation expression, and The static score is adjusted by the score adjustment value. The calculation formula in the present embodiment includes known calculation methods such as four arithmetic operations, factorial calculation, power calculation, logarithmic function, trigonometric function, and the like.

  When the static condition holding unit 120 holds a static condition, the static condition holding unit 120 determines a set of a document ID of a document including a specific node specified by the path expression and a node ID of a specific area specified by the path expression. 1 is created as an index, and a score calculation formula is stored in association with the document ID. The static score calculation unit 144 acquires the document ID of the document including the search character string from the structured document included in the search result information, and creates it as a second index. The static score calculation unit 144 further compares the document IDs of the first index and the second index, and if they match, the value of the node designated as the node ID held in the first index To get. Next, a calculation formula associated with the document ID is acquired, and a score addition value is calculated based on the previously acquired node value. Finally, the static score of the structured document is adjusted by the calculated score addition value.

Here, an example of static score calculation using a job title by the document search apparatus 100 according to the present embodiment is shown. For example, suppose that it is desired to order structured documents in the order of the creator's post when the post classification is as follows.
(Job title, job title) = (President, 1), (Department manager, 10), (Manager, 50), (General employee, 100)
Here, when the structured document in FIG. 7 is to be searched, if the path expression is “position classification = / doc / author [@yakusyoku]”, the value of yakusyoku, which is the attribute of <author>, is the variable “ It can be acquired in "Job title classification". When the score calculation formula is “score = 10000−position classification ^ 2”, search result information as shown in FIG. 8 is displayed. “^” Indicates a power.

As another example, an example of static score calculation using store information by the document search apparatus 100 according to the present embodiment will be shown. For example, it is assumed that the store data includes the values of the east longitude and north latitude of the store position and the values of east longitude and north latitude of the current position.
In this case, from the structured document describing the store information, for example, numerical values of the east longitude and north latitude of the store position are acquired by the following route formula.
Store location East longitude = / doc / shop / gps / tokei
North latitude = / doc / shop / gps / hokui
Further, when the following score calculation formula is used, search result information as shown in FIG. 9 is displayed.

In order to increase the ordering accuracy of the structured documents included in the search result information, it is necessary to acquire various information existing in the structured documents as parameters for score calculation. Furthermore, it is necessary to calculate a score by combining these parameters in a complicated manner.
According to the present embodiment, it is possible to acquire various existing information such as job titles and GPS position information existing in the structured document. In addition, the static score can be calculated using an arbitrary calculation formula using the acquired information. Thereby, the flexibility and accuracy of ordering can be further increased.

Example 9
In this embodiment, the static score is adjusted by a plurality of types of score adjustment values associated with a plurality of types of static conditions, and the static conditions include a plurality of score adjustment values for calculating the static score. A weighting setting for the contribution ratio is included, and the static score calculation unit 144 adjusts the static score by a plurality of score adjustment values based on the weighting setting.

The weighting setting of this embodiment is held in the static condition holding unit 120. The static score calculation unit 144 refers to this weighting setting of the static condition holding unit 120 and calculates a static score.
In the above-described Examples 4 to 7, an example of ordering based on static conditions is shown. These can all be set simultaneously, and a plurality of static conditions can be set simultaneously for the static conditions according to the same embodiment. This may occur, for example, when it is desired to order structured documents based on a plurality of evaluation criteria such as creator, creation date / time, and importance. In this embodiment, it is possible to adjust the weighting of the plurality of static conditions, thereby realizing the optimal ordering according to the common policy of the organization.

  For example, as static conditions in the same structured document, (1) score adjustment by keyword matching according to the fourth embodiment, (2) score adjustment by update date and time according to the sixth embodiment, (3) creation according to the seventh embodiment It is assumed that the score adjustment by the person is set. In this case, if the contribution of (1) is set to 40%, the contribution of (2) is set to 30%, and the contribution of (3) is set to 30%, no matter how the document created by the president, If the keyword does not match, the static score is calculated small. Even if it matches the keyword, if the update date is old, the static score is calculated to be small.

  Note that the weighting setting of this embodiment may be given by the search execution user as a dynamic condition. In this case, the static score calculation unit 144 obtains the weighting setting included in the dynamic condition and calculates the static score. Thereby, it is possible to weight each of the plurality of static conditions based on the subjectivity of the search execution user.

In any of the embodiments herein, the registration of the static condition may be before or after the registration of the structured document. When the static condition is registered before the structured document is registered, the first index described above is created based on the associated static condition when the structured document is registered. When a static condition is registered after registration of a structured document, the associated structured document is scanned to create the first index described above when the static condition is registered.
Further, regarding the association between the static condition and the structured document, the document registration user may select a registered static condition at the time of document registration. Alternatively, a registered static condition may be automatically associated with a registered document based on the document title or creator. Thereby, the labor of the document registration user can be reduced, and mistakes caused by the manual operation of the document registration user can be reduced. Furthermore, it is useful in that the ordering based on the common policy of the organization can be thoroughly implemented.

  Furthermore, in any of the embodiments, the area to which the path expression is designated is not limited to the area of the file that is physically the same as the structured document. For example, it is possible to specify an annotation document storing annotation information associated with an entity document storing entity information shown in FIGS.

  Here, “substance information” is data serving as search target content, such as elements, tags, and attributes. An “entity document” is a structured document that stores entity information. “Annotation information” is data indicating an annotation given by the user to the entity information, such as an element, a tag, and an attribute. An “annotation document” is a structured document that stores annotation information. The entity information and the annotation information are stored separately in separate documents, ie, the entity document and the annotation document, and the correspondence between the data and the document is indexed for each of the entity document and the annotation document. With these two types of index information, the desired entity document can be searched from both the entity information and the annotation information.

  When the destination of the path expression is the annotation information in the annotation document, the entity document can be included in the search result information according to the value of the annotation information, and the dynamic score of the entity document associated with the annotation document and Static score can be adjusted. For example, in the case of the first embodiment, the path expression is set to “// TODO” for the annotation document, and the entity information related to the annotation information in the area of the “TODO” node in the annotation document is used as the search result information. Can be included. Further, for example, in the case of the fourth embodiment, a specific tag in the annotation document is designated, and when the search character string matches the element of the tag, the static document of the entity document related to the annotation document is specified. The target score can be increased and the order can be increased.

  An amendment request from another person for a description of an entity document or a memo to change the description content is often described on the annotation document side. Therefore, the search execution user may want to order the entity information included in the search result information based on the annotation information of the annotation document. In the present embodiment, in response to the needs of the search execution user, the convenience of the search execution user is improved by adjusting the score of the entity document based on the value of the annotation information.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are also within the scope of the present invention. is there.

  In the present embodiment, the XML document has been described as an object. However, the document search apparatus 100 is a document file of a type in which the position of data is specified by a path expression based on a hierarchical structure of tags, such as XHTML, HTML, and SGML. It can be applied to any target.

It is a schematic diagram which shows the outline | summary of a process of the document search device concerning embodiment of this invention. It is a functional block diagram which shows the structure of a document search device. It is a flowchart which shows the flow of a process in a document search device. It is an example of a structured document. It is an example of a structured document. It is a figure explaining the area | region designated with dynamic score area | region designation | designated information. It is an example of a structured document. It is a display example of search result information. It is a display example of search result information.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 Document search apparatus, 110 User interface process part, 112 Static condition reception part, 114 Search request reception part, 116 Search result output part, 120 Static condition holding part, 130 Search execution part, 140 Search ranking control part, 142 Dynamic score calculation unit, 144 Static score calculation unit, 146 Document score assignment unit, 148 Order determination unit, 200 Communication network, 300 Database, 400 Area shown as a whole of structured document 410 Predetermined node of structured document The area indicated by the node in the vicinity of the predetermined node, 412 the area indicated by the node located at the end in the tag structure of the structured document, the step of detecting S1 search request information, the step of acquiring S2 search result information, and the operation of S3 Step of calculating a static score, S4 static score The step of leaving, calculating an S5 document score, determining S6 order of each document, S7 step of the result screen.

Claims (14)

A document search device that searches a database holding a structured document in which a data position is specified by a path expression based on a hierarchical structure of tags, and outputs the search results in order,
A static condition receiving unit for detecting a setting input of a static condition for ordering by the user;
A static condition holding unit for holding the static condition;
A search request receiving unit for detecting input by a user of search request information including a search character string and a dynamic condition for ordering as a search condition of a structured document;
A search execution unit that searches the database for a structured document including the search character string and acquires the search result as search result information;
A dynamic score calculation unit that calculates a degree of matching between a plurality of structured documents included in the search result information and the dynamic condition as a dynamic score;
A static score calculation unit that calculates a degree of conformity between the plurality of structured documents included in the search result information and the static condition as a static score;
A document score assigning unit that calculates a document score for each structured document based on the dynamic score and the static score;
An order determination unit that determines the order of a plurality of structured documents included in the search result information based on the document score assigned to the structured document;
A search result output unit configured to prioritize a plurality of structured documents included in the search result information in the determined order and display them on a screen;
Equipped with a,
The static condition includes a path expression that specifies a specific node in the structured document and a score calculation expression that calculates a score adjustment value based on the value of the specific node.
The static score calculation unit obtains a value of a specific node designated by the path expression from a structured document, calculates a score adjustment value from the value of the specific node and the score calculation expression, and the structure document search apparatus characterized that you adjust the static score document by the score adjustment value. The dynamic condition further includes a weighting setting for a ratio that the dynamic score and the static score contribute in the calculation of the document score,
The document search apparatus according to claim 1, wherein the document score assigning unit calculates the document score from the dynamic score and the static score based on the weighting setting. In the weight setting, when only the static score is weighted and the dynamic score is not weighted,
The document search apparatus according to claim 2, wherein the document score assigning unit invalidates the contribution of the dynamic score to the document score, and calculates the document score based only on the static score. The dynamic condition includes a path expression that specifies a specific area in the structured document,
The search execution unit searches the database for a structured document including the search character string in a specific area specified by the path expression, and acquires the search result as search result information. The document search device according to any one of 1 to 3. The dynamic condition includes a plurality of types of route expressions and score adjustment values for each of the plurality of types of route expressions.
The dynamic score calculation unit adjusts a dynamic score of a structured document including the search character string in an area indicated by the path expression designated as the dynamic condition by a score adjustment value corresponding to the path expression. The document search apparatus according to claim 1, wherein the document search apparatus is a document search apparatus. The dynamic condition includes dynamic score area designation information for designating a dynamic score area as a partial area in the structured document,
The dynamic score calculation unit calculates a dynamic score for the dynamic score area,
The dynamic score area designation information includes an area indicated by a node located at the end in the tag structure of the structured document, an area indicated by a predetermined node of the structured document and a node near the predetermined node, or the structure 6. The document search apparatus according to claim 1, wherein the document search apparatus is information that designates one of the areas indicated as a whole of the digitized document as the dynamic score area. The static condition includes a path expression that specifies a specific area in the structured document, and a static score of the structured document that includes the search character string in the area specified by the path expression. Including additional score adjustment values,
The static score calculation unit further adjusts a static score of a structured document including the search character string in an area specified by the path expression by the score adjustment value. The document search device according to any one of the above. Wherein the static condition, the path expression to specify a particular node of the structured document, the static ordering information indicating the rules for ordering the structured document based on the value of the particular node further Included,
The static score calculation unit obtains a value of a specific node specified by the path expression from a structured document, refers to the static ordering information, and performs the structured based on the value of the specific node. 4. The document retrieval apparatus according to claim 1, further comprising adjusting a static score of the document. The static condition further includes a path expression that specifies a specific node in the structured document and information that specifies a range of values of the specific node.
The static score calculation unit acquires a value of a specific node specified by the path expression from the structured document, and based on the information specifying the value of the specific node and the range, the structured document The document retrieval apparatus according to claim 1, further adjusting the static score of the document. The static condition is associated with a path expression that specifies a specific node in the structured document, a predetermined fixed value to be compared with the value of the specific node, and the predetermined fixed value. Including additional score adjustment values,
The static score calculation unit acquires a value of a specific node specified by the path expression from a structured document, and when the value of the specific node matches the fixed value, the static score of the structured document 4. The document search apparatus according to claim 1, wherein the score is further adjusted by the score adjustment value corresponding to the fixed value. 5. The static score is adjusted by a plurality of types of score adjustment values associated with a plurality of types of static conditions,
The static condition includes a weighting setting for a ratio of a plurality of score adjustment values contributing to the calculation of the static score,
4. The document search apparatus according to claim 1, wherein the static score calculation unit adjusts the static score based on the plurality of score adjustment values based on the weighting setting. 5. The static score is adjusted by a plurality of types of score adjustment values associated with a plurality of types of static conditions,
The dynamic condition includes a weighting setting for a ratio that a plurality of score adjustment values contribute to the calculation of the static score,
The document search apparatus according to claim 1, wherein the static score calculation unit adjusts the static score by a plurality of score adjustment values based on the weighting setting. A document search method for searching a database holding a structured document in which a data position is specified by a path expression based on a hierarchical structure of tags, and ordering and outputting the search results,
Detecting a setting input of a static condition for the ordering by a user;
Detecting input by a user of search request information including a search string and a dynamic condition for ordering as a search condition for a structured document;
Retrieving a structured document including the search character string from the database, and obtaining the search result as search result information;
Calculating a fitness score between a plurality of structured documents included in the search result information and the dynamic condition as a dynamic score;
A static score calculation step of calculating a degree of conformity between the plurality of structured documents included in the search result information and the static condition as a static score ;
Calculating a document score for each structured document based on the dynamic score and the static score;
Determining an order of a plurality of structured documents included in the search result information based on the document score assigned to the structured document;
Prioritizing a plurality of structured documents included in the search result information in the determined order and displaying them on the screen;
Equipped with a,
The static condition includes a path expression that specifies a specific node in the structured document and a score calculation expression that calculates a score adjustment value based on the value of the specific node.
The static score calculation step acquires a value of a specific node specified by the path expression from a structured document, calculates a score adjustment value from the value of the specific node and the score calculation expression, and calculates the structure document search method characterized that you adjust the static score document by the score adjustment value. A computer program for causing a computer to realize a document search function for searching a database holding a structured document whose data position is specified by a path expression based on a hierarchical structure of tags, and ordering and outputting the search results,
A function for detecting a setting input of a static condition for the ordering by a user;
A function of holding the static condition;
A function for detecting input by a user of search request information including a search string and a dynamic condition for ordering as a search condition of a structured document;
A function for retrieving a structured document including the search character string from the database and acquiring the search result as search result information;
A function for calculating a fitness score between a plurality of structured documents included in the search result information and the dynamic condition;
A static score calculation function for calculating a degree of conformity between the plurality of structured documents included in the search result information and the static condition as a static score ;
A function for calculating a document score for each structured document based on the dynamic score and the static score;
A function of determining the order of a plurality of structured documents included in the search result information based on the document score assigned to the structured document;
A function of prioritizing a plurality of structured documents included in the search result information in the determined order and displaying them on a screen;
Is realized on a computer ,
The static condition includes a path expression that specifies a specific node in the structured document and a score calculation expression that calculates a score adjustment value based on the value of the specific node.
The static score calculation function acquires a value of a specific node specified by the path expression from a structured document, calculates a score adjustment value from the value of the specific node and the score calculation expression, and the structure computer program characterized that you adjust the static score document by the score adjustment value.

Images