JP2012212304A - Retrieval result output device, retrieval result output control program, and retrieval result output method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently select an appropriate retrieval result by differentiating the superiority of retrieval results in a case where a plurality of retrieval results are acquired, whose consistency degrees to a retrieval condition are the same in retrieval processing using a database.SOLUTION: A retrieval result output device detects, at a time of completing retrieval of attribute information storage tables storing attribute information consistent with a retrieval condition, upper nodes without duplication, positioned upper by one layer than nodes corresponding to the attribute information storage tables storing the attribute information consistent with the retrieval condition, counts, for each of the upper nodes, the number of nodes corresponding to the attribute information storage tables storing the attribute information consistent with the retrieval condition among lower nodes positioned lower by one layer than the detected upper nodes, rearranges the order of the upper nodes in the descending order based on the number of the counted lower nodes, and outputs the retrieval result according to the order.

Description

  In the present invention, when a plurality of search results having the same degree of match with the search condition are obtained in the search processing using the database, another criterion is applied to output the search results with superiority or inferiority. The present invention relates to a search result output device, a search result output control program, and a search result output method.

  In a search result output device such as a directory server or a service server that provides a search function for functional groups and individuals belonging to the functional groups, the structure of the directory tree for organizing functional group and individual attribute information is the organizational structure of the functional group. In general, it has a layered structure.

  When searching for functional group or individual information using a search result output device such as a directory server or a service server, a search method and search conditions that follow a directory tree arranged in a hierarchical structure following the organizational structure There is a method of searching by specifying a keyword.

  For example, when searching for individuals who are familiar with a certain field, a directory tree structure representing an organizational structure, that is, a directory tree structure formed by using the position occupied by each individual in the organizational structure of a functional group as a reference for attribute information It becomes necessary to perform a search using different attribute information different from each other, that is, attribute information such as whether or not the user is familiar with a certain field.

  Here, it is determined whether or not the individual is familiar with a certain field by referring to the attribute information storage table of each individual according to the permutation of each node of the directory tree, that is, each individual's position while tracing the directory tree. If attribute information such as the names of individuals who are familiar with a certain field are output as search results in order, the output order is simply in the order of the hierarchical structure of the directory tree.

  In addition, ignoring the hierarchical structure of the directory tree, it is determined whether each individual is familiar with a certain field by referring to the attribute information storage table of each individual from one end, and is familiar with a certain field. If attribute information such as the name of an individual is sequentially output as a search result, the output order is simply the order according to the registration order of the attribute information storage table and the alphabetical order of the name of each individual.

  That is, in any case, the output order only reflects the hierarchical structure of the directory tree unrelated to the search condition, the registration order of the attribute information storage table of each individual, and the like.

  Therefore, when a plurality of search results having the same degree of match with the search condition are output, it is not possible to give a superiority or inferiority which individual is particularly familiar with a certain field.

  As a result, it is not easy to identify the most appropriate person asking a question about a certain field, and it is necessary to take time and effort to actually ask some individuals whose names are output as search results. There is a case. Accordingly, it is desirable that the search results that are more appropriate, that is, in this case, the names of individuals who are particularly familiar with a certain field are output first.

  As an apparatus for optimizing the output of search results, as disclosed in Patent Document 1, in addition to keywords as search conditions, the depth in the directory tree is further used as another criterion to weight the results. What is to be done is known, but this only uses the depth from the root of the directory tree and does not take into account the relative relationship between the search results to be output. It is not possible to specify which of the results is most desirable.

Further, Patent Document 2 summarizes information such as whether individual attribute information such as C ⁺⁺ is familiar with each team, and further, the number of individuals having the same attribute information within the same team, such as C ^++. A search system has been proposed in which the number of individuals familiar with the field is managed in a team profile database. However, it is necessary to create an appropriate team profile in advance and register it in the database. This is not a software resource that determines the basic structure of the database, that is, the structure of the directory tree. There is no technical idea that uses the output order of search results to identify the importance of search results.

Japanese Laid-Open Patent Publication No. 2003-186912 (paragraph 0021) Japanese Laid-Open Patent Publication No. 2005-208945 (paragraph 0043, FIG. 6)

  An object of the present invention is to provide a directory tree structure, which is the basis of a database structure, as another criterion when a plurality of search results having the same degree of match with a search condition are obtained in a search process using a database. To provide a search result output device, a search result output control program, and a search result output method capable of distinguishing superiority and inferiority of search results by using in combination.

The search result output apparatus according to the present invention includes a node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and the node information storage table corresponding to each of the nodes. An attribute information storage table storing attribute information other than the reference attribute information is connected to a storage device previously stored as a database, and the attribute information storage table in the database is searched based on a search condition input from the interface. A search result output device for outputting, as a search result, attribute information of a type specified as an output target among attribute information stored in an attribute information storage table storing attribute information that matches the search condition, In particular, to achieve
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Upper node detecting means for detecting nodes without duplication referring to the node information storage table;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means to perform,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means;
Search the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node according to the order of the upper nodes rearranged by the sorting means. And a search result output means for outputting the result.

The search result output control program of the present invention is a node that stores node specifying information for specifying each node of a directory tree that represents a hierarchical structure of reference attribute information for each node in order to achieve the same object as described above. A computer connected to a storage device that stores an information storage table and an attribute information storage table storing attribute information other than the reference attribute information corresponding to each of the nodes in advance as a database;
Search means for searching an attribute information storage table storing attribute information matching the search condition from the attribute information storage table in the database based on a search condition input from an interface;
When the search by the search means is completed, an upper node located one level higher than the node corresponding to the attribute information storage table storing the attribute information matching the search condition is duplicated with reference to the node information storage table Upper node detecting means for detecting without,
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means; and
Of the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node in the order of the upper nodes rearranged by the sorting means It has a configuration characterized by functioning as search result output means for outputting attribute information of a type designated as an output target as a search result.

The search result output method of the present invention corresponds to a node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node and each of the nodes. An attribute information storage table storing attribute information other than the reference attribute information is stored in advance as a database, the attribute information storage table in the database is searched based on a given search condition, and the search A search result output method for outputting, as a search result, attribute information of a type designated as an output target among attribute information stored in an attribute information storage table storing attribute information that matches a condition. To achieve
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Detecting nodes with reference to the node information storage table without duplication;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the detected upper node is counted for each upper node,
Rearranging the rank of the upper node in descending order based on the number of counted lower nodes,
The attribute information stored in the attribute information storage table storing the attribute information that matches the search condition among the lower nodes positioned one hierarchy lower than each upper node in accordance with the rearranged upper node order is output as the search result. It has the structure characterized by this.

The search result output device, the search result output control program, and the search result output method of the present invention are configured to retrieve attribute information that matches the search condition when the search of the attribute information storage table that stores the attribute information that matches the search condition is completed. Referring to the node information storage table, the upper node located one level higher than the node corresponding to the stored attribute information storage table is detected without duplication, and the lower node located one hierarchy lower than the detected upper node is searched. Count the number of nodes corresponding to the attribute information storage table that stores attribute information that matches the conditions for each upper node, and rearrange the rank of the upper node in descending order based on the number of counted lower nodes In accordance with the order of the rearranged upper nodes, the attribute information that matches the search condition among the lower nodes located one hierarchy lower than each upper node is stored. Since the attribute information of the type specified as the output target among the attribute information stored in the sex information storage table is output as the search result, the content and number of the attribute information output as the search result is a general search. Same as in the case of the result output device, search result output control program, and search result output method, but with the hierarchical structure of the attribute information as a reference as another reference, it matches the search condition within the same hierarchy in this hierarchical structure The attribute information that is the search result is output in order from the hierarchy having the largest number of nodes corresponding to the attribute information storage table storing the attribute information.
Therefore, even when multiple search results with the same degree of match with the search conditions are obtained, the superiority of the search results can be differentiated by checking the output order of the search results, and the search results that are more appropriate Can be easily obtained.

For example, in the configuration where the structure of the directory tree is a hierarchical structure that follows the organizational structure of the functional group, and the database is created as attribute information based on the position occupied by each individual in the organizational structure, that is, a directory tree node, the name of each individual When the information on whether or not the individual is familiar with a certain field is stored in the attribute information storage table for each node as other attribute information, the fact that the person is familiar with the certain field is used as a search condition. It is assumed that an attribute information storage table is searched and an individual name is output as attribute information of a search result.
At that time, even if the name of an individual who is familiar with a certain field is extracted by referring to the attribute information storage table while tracing the directory tree, the hierarchical structure of the directory tree is ignored and each node, that is, each node Even if the names of individuals who are familiar with a certain field are extracted by referring to the attribute information storage table for each individual from one end, some output results, that is, the names of some individuals, However, in the present invention, in addition to the search condition, the directory tree structure that is the basis of the database structure is used as another criterion, for example, the directory tree representing the organizational structure. By using a hierarchical structure together, it is positioned one level higher than the node corresponding to the attribute information storage table that stores individuals familiar with a certain field. In other words, each of the upper nodes having at least one node corresponding to the attribute information storage table storing an individual familiar with a certain field in the hierarchy immediately below is familiar with a certain field. Count the number of nodes corresponding to the attribute information storage table storing the individual in the hierarchy immediately below it, and the node corresponding to the attribute information storage table of the individual familiar with a certain field in the descending order of the number. The ranks of the upper nodes are rearranged in descending order, and the names of individuals as search results are output in order from the lower hierarchy of the first hierarchy according to the rank of the upper node.
In other words, as a result, the names of individuals who are search results are output in the order of the hierarchy in which there are many individuals who are familiar with a certain field in the same hierarchy. The name of a person who is particularly familiar with the field is likely to be output as the search result of the first group, and it is expected that multiple search results with the same degree of match with the search conditions will be effectively differentiated. The

It is the functional block diagram which simplified and showed the structure of the search result output device of one Embodiment to which this invention is applied. It is the conceptual diagram which showed an example of the structure of a directory tree. It is the conceptual diagram which showed an example of the logical structure of a node information storage table. It is the conceptual diagram which showed an example of the logical structure of an attribute information storage table. It is the conceptual diagram which showed the example of a connection at the time of using a search result output device via a network. It is the conceptual diagram which showed an example of the logical structure of a keyword corresponding | compatible search result storage table. It is the conceptual diagram which showed an example of the logical structure of a count-up table. It is the conceptual diagram which showed an example of the logical structure of a sorting table. It is the conceptual diagram which showed an example of the logical structure of a search result storage table. It is the block diagram shown about the outline of the hardware constitutions of a search result output device. It is the flowchart shown about the outline of a structure of a search result output control program. It is the continuation of the flowchart shown about the outline of the structure of the search result output control program. It is the continuation of the flowchart shown about the outline of the structure of the search result output control program. FIG. 14A is a diagram conceptually showing an output example of a search result, FIG. 14A shows an output example when the present invention is applied, and FIG. 14B shows a directory tree structure as a basis of a database structure. An example of output when not used together is shown.

  Next, an example is given about the form for implementing this invention, and it demonstrates concretely with reference to drawings.

  FIG. 1 is a functional block diagram schematically showing the configuration of a search result output device 1 and its storage device 2 according to an embodiment constituted by a computer such as a personal computer or a workstation.

  The storage device 2 connected to the search result output device 1 includes a node information storage table storing node specifying information for specifying each node of the directory tree representing the hierarchical structure of the reference attribute information, and each of the nodes. The attribute information storage table storing attribute information other than the reference attribute information corresponding to is stored as the database 3 in advance.

  An example of the structure of the directory tree in the database 3 is shown in the conceptual diagram of FIG. Here, as an example, a hierarchical structure of a directory tree constructed following the organizational structure of a functional group such as a company is shown. Each of the nodes P1 to P46 represents a position occupied by each individual in the organizational structure of the functional group, such as a post. That is, the attribute information used as a reference in this example is the position that each individual occupies in the organizational structure, and the directory tree in FIG. 2 is a directory tree that represents the relationship of the positions that each individual occupies with nodes P1 to P46. become.

  The node specifying information required for specifying each of the nodes P1 to P46 is stored in a node information storage table provided for each node and stored in the database 3. An example of the logical configuration of the node information storage table is shown in FIG. FIG. 3 is an example of a node information storage table that identifies the node P14 in the directory tree shown in FIG. In this node information storage table, a pointer (hereinafter referred to as an upper node pointer) for specifying a node directly above the node and a pointer to a previous node having the same upper node (hereinafter referred to as the previous node) Pointer), a pointer to a subsequent node having the same upper node (hereinafter referred to as a subsequent node pointer), and a pointer (hereinafter referred to as “header node”) for identifying the first node among nodes positioned directly below the node. The same as a pointer (hereinafter referred to as an actual data pointer) for specifying the location of an attribute information storage table that stores attribute information other than reference attribute information corresponding to the node) Stores a pointer (hereinafter referred to as a relative identification pointer) representing a relative identification name between nodes having upper nodes.

  As apparent from FIG. 2, taking the node P14 as an example, the upper node pointer is a pointer representing the node P7, and the previous node pointer is a pointer representing the node P13. The rear node pointer is a pointer representing the node P15, and the lower head node pointer is a pointer representing the node P26. The node information storage table includes all information necessary for specifying a directly upper layer, a directly lower layer, and adjacent nodes in the same layer with respect to the node corresponding to the node information storage table. Therefore, the directory tree as shown in FIG. 2 can be constructed by referring to all the node information storage tables of the nodes P1 to P46. That is, the information necessary for actually specifying the structure of the directory tree is only the node information storage table for each node, and the node information storage table corresponding to each of the nodes P1 to P46 is collectively stored in the database 3. It is stored. As shown in FIG. 3, the position that is the reference attribute information, such as a position, may be specified in the node information storage table.

  Further, FIG. 4 shows an example of the attribute information storage table corresponding to the node P14 as an example of the attribute information storage table. As already described, the attribute information storage table is for storing attribute information other than the reference attribute information corresponding to each of the nodes, that is, other attribute information excluding the personal status in this example, For example, attribute information such as the name of the individual, the department to which the individual belongs, the years of service of the individual, the field familiar to the individual, the skill of the individual in the field, and the like are stored. The name is a kind of attribute information for identifying each individual. The attribute information storage table corresponding to each of the nodes P1 to P46 is also collectively stored in the database 3 as in the case of the node information storage table described above. The attribute information stored in the attribute information storage table can be used as a search condition either alone or in combination. For example, it is possible to output personal information that is somewhat familiar with the field as a search condition, that is, the name as attribute information that is the search result, and further, the AND condition of the field and skill is used as the search condition. It is also possible to output personal information that is extremely familiar with the field, that is, the name as attribute information as a search result. Such search processing is already known, and in any case, a plurality of search results, that is, names may be output.

  On the other hand, as shown in FIG. 1, the main part of the search result output apparatus 1 composed of a computer such as a personal computer or a workstation has an interface A for inputting search conditions and a search condition input from the interface A. Based on the attribute information storage table of the database 3 based on the attribute information storage table storing attribute information that matches the search condition, the search means B matches the search condition when the search by the search means B is completed. An upper node detecting means C for detecting an upper node located one level higher than a node corresponding to the attribute information storage table storing the attribute information with reference to the node information storage table, and an upper node detecting means C. Attributes that match the search conditions among the lower nodes located one layer below the higher node The counting means D for counting the number of nodes corresponding to the attribute information storage table for storing information for each upper node, and the upper node detecting means C based on the number of lower nodes counted by the counting means D. Sorting means E for rearranging the ranks of the higher nodes in descending order, and an attribute that matches the search condition among lower nodes positioned one hierarchy lower than each upper node according to the order of the upper nodes rearranged by the sorting means E Of the attribute information stored in the attribute information storage table storing the information, it is constituted by search result output means F for outputting the attribute information of the type specified as the output target, for example, the name as the search result.

  The interface A for inputting the search condition is a keyboard or the like provided in the search result output device 1 when the search result output device 1 is used alone. Further, when the search result output device 1 is connected to a network 4 such as the Internet or an in-house LAN as shown in FIG. 5 and used from other client machines 5, 5, 5,. The communication interface provided in the search result output device 1 functions as the interface A in FIG.

Regarding the search means B, its configuration and function are already known. That is, the search means B traces the directory tree of the database 3 and refers to the contents of the attribute information storage table of each individual according to each node of the directory tree, that is, the permutation of the status of each individual. ^It is determined whether or not it matches the field such as ^++, and the attribute information storage table whose attribute information matches the search condition is searched. Alternatively, the search means B ignores the hierarchical structure of the directory tree and refers to each person's attribute information storage table from one end to determine whether each person's attribute information matches a field such as C ^++. The attribute information storage table whose attribute information matches the search condition may be searched. Further, the search condition may be specified as a single unit or a combination of a plurality of search conditions. These points are the same as general search processing.

  However, this search means B only searches the attribute information storage table whose attribute information matches the search condition, and immediately retrieves the search result, that is, the attribute information of the type specified as the output target, such as the personal name. It does not output externally.

  The search means B in this embodiment searches and obtains an attribute information storage table whose attribute information matches the search condition, and each time the attribute information storage table whose attribute information matches the search condition is found, the attribute information storage table The value of the node corresponding to is stored in the keyword correspondence search result storage table as shown in FIG.

The values ₁ to n ₁ shown in FIG. 6 are the addresses of the records in the keyword-corresponding search result storage table, and a _{1 to an n1} are the records corresponding to the addresses ₁ to n ₁ , and the attribute information obtained by the search The value of the node corresponding to the storage table is stored. When searching the attribute information storage table of each individual according to the permutation of each person's status while following the directory tree, or when referring to the attribute information storage table of each individual from one end while ignoring the directory tree hierarchy There is a difference in the order of storing the node values corresponding to the attribute information storage table matching the conditions in the keyword correspondence search result storage table, but the combination of the node values stored in the keyword correspondence search result storage table is any This is the same even when this method is applied.

  The upper node detecting means C is the attribute information that matches the search condition when the search process by the search means B is completed, that is, when the search means B finishes writing the value of the node in the keyword correspondence search result storage table. The node corresponding to the attribute information storage table storing the information, that is, the upper node positioned one level higher than the node stored in the keyword correspondence search result storage table of FIG. 6 is detected without reference to the node information storage table. Then, the value of the upper node is stored in a count-up table as shown in FIG.

Thus, for example, if a node P14 shown in FIG. 2 to record a ₁ of the first address in the keyword corresponding search result storage table in Fig. 6 has been stored, the first address in the count-up table of Figure 7 records b ₁ , the node P7 directly above the node P14 is stored.
Similarly, if a node P8 shown in Figure 2 the keyword corresponding search results second address in the memory table records a ₂ in FIG. 6 has been stored, the count-up second address of record in the table b ₂ in FIG. 7 Stores the node P5 directly above the node P8.
Furthermore, if a node P15 shown in FIG. 2 the keyword corresponding search results third address record a ₃ in the storage table in FIG. 6 has been stored, the node of the direct higher is it comes to node P7, the since already upper node P7 in step is written to the first address of the record b ₁ in the count-up table of Figure 7, a third address of the record in the count-up table of the parent node detecting means C is 7 overlap again It is not possible to write the node P7 to b _3.
Therefore, the number n2 of nodes stored in the count-up table in FIG. 7 is either equal to or smaller than the number n1 of nodes stored in the keyword correspondence search result storage table in FIG.

  The counting means D stores the attribute information that matches the search condition among the upper nodes detected by the upper node detecting means C, that is, the lower nodes located one layer lower than the upper node stored in the count-up table. The number of nodes corresponding to the attribute information storage table is counted for each upper node, and the counted value is stored in the count-up table in association with the upper node.

Thus, for example, if the node P7 shown in FIG. 2 to record b ₁ first address in the count-up table of Figure 7 has been stored, the record c ₁ of the first address in the count-up table in Fig. 7 Of the nodes P13, P14, and P15 directly below the node P7, the number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition, that is, one of the values 1 to 3 is written. It will be.
Since the upper node is one that has been detected attribute information matching the original search criteria as the upper node of the node corresponding to the attribute information table storing, impossible be counted value stored in the record c ₁ becomes 0 Absent. The same applies to the records c _{2 to} c _n2 .

The sorting means E is based on the magnitude of the number of lower nodes counted by the counting means D, that is, based on the magnitudes of the count values c _{1 to} c _n2 stored in the first address to the n2th address in the count-up table of FIG. The upper nodes detected by the upper node detecting means C, that is, the upper nodes b _{1 to} b _n2 stored in the first address to the n2 address in the count-up table of FIG. 7 are arranged in descending order. instead, the write in order to the first address, second n2 address record _d 1 _{to d n2} sorting table as shown in FIG.

Therefore, for example, if the maximum value among the count values c _{1 to} c _n2 stored in the count-up table is c _i2 and the minimum value is c ₂ , the count value is set to c _{i2 in} the count-up table. The value of the corresponding node b _i2 is written in the record d ₁ of the first address of the sorting table shown in FIG. 8, and the value of the node b ₂ corresponding to the value of c _{2 in} the count-up table is shown in FIG. It will be written to a n2 address record d _n2 sorting table.

The search result output means F _first sorts each upper node according to the order of the upper nodes rearranged by the sorting means E, that is, according to the order of the upper nodes written in the records d _{1 to} d _n2 of the sorting table of FIG. Attribute information of the type specified as the output target among the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one layer lower than d _{1 to} d _n2 The search result storage table as shown in FIG. 9 is stored as a search result corresponding to each of the upper nodes d _{1 to} d _n2 .

For example, on the premise that the name of the attribute information stored in the attribute information storage table is output as the final search result, the fact that the attribute information of the field is C ⁺⁺ is input to the interface A as a search condition. FIG. 2 shows a record of the upper node d ₁ of the search result storage table as shown in FIG. 9 as a result of executing the above-described processes by the search means B, upper node detection means C, counting means D, and sorting means E. Assume that the indicated node P6 is stored. Further, C ⁺⁺ is stored in the field of the attribute information field in the attribute information storage table corresponding to the lower node P10 located one layer lower than the node P6 shown in FIG. 2, and one node lower than the node P6. C ⁺⁺ is also stored in the field of attribute information field in the attribute information storage table corresponding to the lower node P11 located in the attribute information storage table, while in the attribute information storage table corresponding to the lower node P12 located one level lower than the node P6. It is assumed that COBOL is stored in the attribute information field.
In this case, among the lower nodes P10 to P12 located one layer lower than the node P6, the lower nodes that match the search condition that the field is C ⁺⁺ are P10 and P11. Therefore, as shown in FIG. The search result storage table corresponds to the record of the upper node d ₁ , and corresponds to the name e _{1 used} in the name field of the attribute information of the attribute information storage table corresponding to the lower node P 10, and the lower node P 11. The name e ₂ stored in the name column of the attribute information in the attribute information storage table is stored.

Furthermore, the search result output means F of this embodiment has a distribution density calculation function, and the value of [number of lower nodes matching the search condition / total number of lower nodes] corresponds to each name as a search result. Then, it is stored as a distribution density in a search result storage table as shown in FIG.
Taking the above-mentioned relationship between the high node P6 and lower nodes P10~P12 and name _e 1, _{e 2} as an example, the value _{f 1} of the distribution density corresponding to name _{e 1} is [2/3] = 0.66 Further, The value f _{2 of the} distribution density corresponding to the name e ₂ is also [2/3] = 0.66.

  Then, the search result output means F finally displays the names stored in the search result storage table as shown in FIG. 9 and the value of the distribution density corresponding to each name as a set of data as shown in FIG. Are output to the outside as search results in order from the first record in the search result storage table.

Accordingly, in the example of FIG. 9, the output order of the final search results is [name e ₁ , distribution density f ₁ ], [name e ₂ , distribution density f ₂ ], [name e ₃ , distribution density f]. ₃ ], [name e ₄ , dense distribution f ₄ ],...

  Here, as an example, a case is illustrated in which a field is used as a search condition among attribute information stored in the attribute information storage table, and a name is output as a final search result among attribute information stored in the attribute information storage table. However, other attribute information stored in the attribute information storage table such as years of service may be used as a search condition, or a plurality of attribute information such as field and skill AND conditions may be used as a search condition. Or, if attribute information such as employee number is stored in the attribute information storage table as attribute information for identifying each individual, the employee number etc. is searched for, that is, the attribute information of the type to be output It is also possible to specify as.

  FIG. 10 is a block diagram showing an outline of a hardware configuration of the search result output apparatus 1 configured by a computer such as a personal computer or a workstation.

  As shown in FIG. 10, the search result output device 1 composed of a computer includes a microprocessor 6 for arithmetic processing, a ROM 7 storing a basic control program required for driving control of the microprocessor 6, and Communication used for connection between the nonvolatile memory 8 for storing various parameters, the RAM 9 used for temporary storage of calculation data, the hard disk drive 10 functioning as a mass storage device, and the network 4 Interface 11 and an external drive 12 for reading a program or data from a non-temporary recording medium such as a DVD, a CD, a flash memory, and the like. Keyboard 14 and mouse 15 functioning as an interface and display Ray 16 is connected.

  As described above, when the search result output device 1 is used alone, the keyboard 14 of the search result output device 1 functions as the interface A in FIG. 1, and the search is performed as shown in FIG. When the result output device 1 is connected to the network 4 and used from other client machines 5, 5, 5,... Connected to the network 4, the communication interface 11 of the search result output device 1 is used. Functions as the interface A in FIG.

  Further, the large-capacity hard disk drive 10 stores the storage device 2 in FIG. 1, that is, the node information storage table for each node (for example, see FIG. 3) and the attribute information storage table for each node (for example, see FIG. 4) for the number of nodes. It also functions as a means for storing the database 3 gathered by the amount of.

Further, the hard disk drive 10 causes the microprocessor 6 of the search result output device 1 to function as the search means B, upper node detection means C, counting means D, sorting means E, and search result output means F in FIG. A search result output control program is stored in advance.
The search result output control program may be installed using a non-temporary recording medium on which the search result output control program is recorded, such as a DVD, CD, flash memory, etc., and the external drive 12.

  Further, a part of the storage area of the RAM 9 includes the keyword-corresponding search result storage table (see FIG. 6), the count-up table (see FIG. 7), the sorting table (see FIG. 8), and the search result storage table (see FIG. 9). ) Etc. also function as a table generation area for generating temporarily.

  FIGS. 11 to 13 are flowcharts showing an outline of the configuration of a search result output control program that causes the microprocessor 6 to function as the search means B, the upper node detection means C, the counting means D, the sorting means E, and the search result output means F. is there.

  Next, the overall processing operation, search result output control program, and search result output method of the search result output apparatus 1 according to this embodiment will be specifically described with reference to the flowcharts of FIGS.

Here, as an example, a case will be described in which a field is used as a search condition among attribute information stored in the attribute information storage table, and a name is output as a final search result among the attribute information stored in the attribute information storage table. To do. The search condition is C ⁺⁺ .

When it is confirmed that a search request including one or more search conditions is input to the search result output device 1 via the keyboard 14 functioning as the interface A or the communication interface 11 (step S1), first, the microprocessor 6 However, the protocol of the search request is analyzed, and search conditions such as C ⁺⁺ are extracted from the search request (step S2).

Then, the microprocessor 6 functioning as the search means B sequentially searches the attribute information storage table for each of the nodes P1 to P46 stored in the database 3 stored in the hard disk drive 10 functioning as the storage device 2, and gives it. An attribute information storage table storing the obtained search condition C ⁺⁺ as field attribute information is obtained (step S3), and the value of the node corresponding to the attribute information storage table is a keyword correspondence search result storage table as shown in FIG. Are stored in time series in the search order (step S4).

In the example of FIG. 6, nodes corresponding to the attribute information storage table in which C ⁺⁺ as the search condition is stored as field attribute information are represented by a _{1 to} a _n1 .

  As described above, it is possible to determine whether or not the content of the attribute information matches the search condition by referring to the content of the attribute information storage table according to the permutation of the node while following the directory tree of the database 3, or It may be determined whether or not the content of the attribute information matches the search condition by ignoring the hierarchical structure of the directory tree and referring to the attribute information storage table from one end. The processing from step S1 to step S4 is known as general search processing.

Next, the microprocessor 6 functioning as the upper node detecting means C once initializes the value of the index i used for searching the keyword-corresponding search result storage table of FIG. 6 to 0 (step S5), and again sets the value of the index i. Is incremented by 1 (step S6), and it is determined whether the current value of the index i is within the range of the total number n1 of nodes stored in the keyword correspondence search result storage table (step S7). Incidentally, since the value of the index i is equivalent to the value of the address i1 in the FIG. 6, the i replaced with i1 is to refer to FIG. 6 from the description of the flowchart, and shall be read with a _i and a _i1 To do.

Here, if the current value of the index i is within the range of the total number n1 of nodes stored in the keyword correspondence search result storage table of FIG. 6 and the determination result in step S7 is true, the upper node is detected. This means that the node a _i to be left still remains in the keyword correspondence search result storage table as shown in FIG.

Therefore, in this case, the microprocessor 6 which serves as the upper node detecting means C accesses the database 3, node information storage table corresponding to the keyword corresponding search result record a _i node of the i address in the memory table of FIG. 6 refers to the upper node pointer, determine the upper node _{b i} node _{a i} (step S8).

Next, the microprocessor 6 functioning as the upper node detecting means C refers to the count-up table as shown in FIG. 7 and avoids inadvertent extraction processing, so that the upper node b _i obtained in step S8 is processed. Is already stored in the count-up table (step S9).

Then, the microprocessor 6 functioning as the upper node detecting means C only in the case where the upper node b _i obtained in the process of step S8 is not yet stored in the count-up table of FIG. the upper node b _i obtained by the processing by adding the detection order of the count-up table as shown in FIG. 7 is stored (step S10), and the index k representing the number of upper node stored in the count-up table The value is incremented by 1 (step S11).

Next, the microprocessor 6 functioning as the counting means D is newly assigned to the upper node b _i newly stored in the count-up table in FIG. 7, that is, the lower node located one layer lower than the upper node obtained in the process of step S8. The value of the index j for selecting one lower node from which information is to be read out of the nodes is once initialized to 0 (step S12), and the value of the index j is incremented by 1 again (step S13).

Then, the microprocessor 6 which functions as counting means D accesses the database 3, and the lower head node pointer of the node information table corresponding to the upper node b _i obtained by the processing in step S8, the lower head node pointer Referring to the subsequent node pointer of the node information storage table corresponding to the node specified by the node, and the subsequent node pointer of the node information storage table corresponding to the node specified by the node pointer, etc. step selects the lower node located on the j-th from the top among the nodes located one level lower upper node b _i obtained by the processing in S8, with reference to the attribute information storage table corresponding to this node ( step S14), and located one level lower upper node _{b i} obtained by the processing in step S8 Attribute information field in the attribute information storage table corresponding to the lower node located in the j-th from the head of that node determines whether match the C ⁺⁺ is a search condition (step S15).

Then, if the judgment result of step S15 is Yes, that is, corresponding to the lower node located in the j-th from the top among the nodes located one level lower upper node b _i obtained by the processing in step S8 The microprocessor 6 functioning as the counting means D is newly added as shown in FIG. 7 only when it is clear that the field attribute information in the attribute information storage table matches the search condition C ⁺⁺ . The count value c _i corresponding to the upper node b _i stored in the count-up table is incremented by 1, and the current value is updated to the count-up table as shown in FIG. 7 in correspondence with the upper node b _i . (Step S16).

Next, the microprocessor 6 functioning as the counting means D determines whether or not the subsequent node pointer is stored in the node information storage table corresponding to the jth lower node selected in the process of step S14. , the j-th lower node selected in the process of step S14 determines whether was the last of the node located one level lower upper node b _i (step S17).

If the determination result in step S17 is false, the subsequent node pointer is stored in the node information storage table corresponding to the jth lower node selected in the process in step S14. , that is, since the node located one level lower upper node b _i obtained by the processing in step S8 means that there are other, the microprocessor 6 which functions as counting means D is again the process of step S13 Then, based on the updated current value of the index j, the processes in steps S13 to S17 are repeatedly executed in the same manner as described above.

Therefore, finally, when the decision result in the step S17 is Yes, all the nodes located one level lower upper node stored in the count-up table b _i, the attribute information storage corresponding to the node The table is referred to, and it is determined whether or not the attribute information in the field of the attribute information storage table matches the search condition C ^++, and the count value c _i corresponding to the node is determined according to the determination result. The process of updating the value and re-storing it in the count-up table is completed.

That is, when the determination result of step S17 is Yes, and C ⁺⁺ is one of all the attribute information search fields in the attribute information table of the node located one level lower upper node b _i The total number of matching nodes is stored in a count-up table as shown in FIG. 7 as a count value c _i corresponding to the upper node b _i .

  In this way, the process of counting the number of nodes in which the attribute information in the field in the attribute information storage table matches the search condition among all the nodes located one hierarchy lower with respect to one upper node stored in the count-up table Then, the microprocessor 6 proceeds to the process of step S6 again, and repeatedly executes the processes from step S6 to step S17 in the same manner as described above based on the updated current value of the index i.

Therefore, finally, when the determination result in step S7 becomes false, all of the upper nodes stored in the count-up table of FIG. 7, for example, each of the upper nodes b _{1 to} b _n2 correspond to the upper nodes. The writing process of the counted values c _{1 to} c _n2 is completed.

In this embodiment, each time the microprocessor 6 functioning as the upper node detecting means C detects the upper node and stores it in the count-up table of FIG. 7 (see step S6 to step S10), the micro functioning as the counting means D. The processor 6 obtains the number of nodes whose attribute information in the field in the attribute information storage table matches the search condition C ⁺⁺ out of all the nodes located one layer lower than the upper node and associates it with the upper node. 7 is stored in the count-up table in FIG. 7 (see Steps S11 to S17), but the microprocessor 6 functioning as the higher-level node detection means C detects all the higher-level nodes and stores them in the count-up table in FIG. After storing, all nodes located one level lower for each higher node Among them, the number of nodes whose attribute information in the field in the attribute information storage table matches the search condition C ⁺⁺ may be obtained and stored in the count-up table in FIG. 7 in association with each upper node.

Next, the microprocessor 6 functioning as the sorting means E refers to the count-up table as shown in FIG. 7, and the magnitude of the lower nodes counted by the counting means D, that is, in the count-up table of FIG. Based on the magnitudes of the count values c _{1 to} c _k stored in the first address to the k-th address, the higher-order node detected by the higher-order node detection means C, that is, the first address to the second address in the count-up table of FIG. It sorted k the order of address to the stored upper node b ₁ ~b _k in descending order, the record d ₁ to d _k of the first address, second k address of such sorting table as shown in FIG. 8 Write in order (step S18). Note that the value of the index k indicating the number of upper nodes stored in the count-up table, more specifically, the value (final value) of the index k after the point when the determination result in step S7 becomes false is shown. since it corresponds to the value of the address n2 in the 7 and 8, the k replaced and n2 to refer to FIGS. 7 and 8 from the description of the flowchart, to be replaced with d _k and d _n2. Further, since simple sorting processing is known, it will not be described in particular here.

  Next, the microprocessor 6 functioning as the search result output means F temporarily initializes the value of the index i used for searching the sorting table to 0 (step S19), and once again increments the value of the index i by 1 ( Step S20), it is determined whether or not the current value of the index i is within the range of the total number k of the upper nodes stored in the sorting table (Step S21).

Here, if the current value of the index i is within the range of the total number k of the upper nodes stored in the sorting table of FIG. 8 and the determination result in step S21 is true, the attribute information that matches the search condition This means that the upper node d _i that should detect the lower node having the attribute information storage table that stores “” still remains in the sorting table as shown in FIG.

Therefore, in this case, the search result output unit microprocessor 6 which functions as F is one read from the i address sorting table based on the current value of an upper node d _i of the index i (step S22), and then, the upper corresponding attribute among the lower node located one level lower value and the upper node d _i of the index j for selecting the lower nodes to be read information from the lower node located one level lower node d _i The value of the count index C for counting the number of lower nodes whose attribute information in the field in the information storage table matches the search condition C ⁺⁺ is once initialized to 0 (step S23). Is incremented by 1 (step S24).

Then, the search result output unit microprocessor 6 which functions as F accesses the database 3, and the lower head node pointer of the node information table corresponding to the upper node d _i read in the processing in step S22, the lower the top Refers to the subsequent node pointer of the node information storage table corresponding to the node specified by the node pointer, and the subsequent node pointer of the node information storage table corresponding to the node specified by the node pointer. The j-th lower node from the head is selected from the nodes located one layer lower than the upper node d _i read in step S22, and the attribute information storage table corresponding to this node is referred to. Te (step S25), and the upper node d read in the processing of step S22 Attribute information field in the attribute information storage table corresponding to the lower node located in the j-th from the top among the nodes located one level lower it is judged whether or not consistent with C ⁺⁺ as the search condition ( Step S26).

Then, if the judgment result of step S26 is Yes, that is, the lower is located in the j-th from the top among the nodes located one level lower upper node d _i read in the processing in step S22 in this time Only when it becomes clear that the attribute information of the field in the attribute information storage table corresponding to the node matches the search condition C ⁺⁺ , the microprocessor 6 functioning as the search result output means F is subordinate to this. It reads the attribute information of the attribute information, i.e. name of the specified type as an output object to be search results from the attribute information storage table corresponding to the node, as shown in FIG. 9 in correspondence with this name to the upper node d _i order is stored in the sorting already search result storage table (step S27), and the lower located one level lower upper node d _i Among the rank nodes, the value of the count index C for counting the number of lower nodes whose attribute information in the corresponding attribute information storage table matches the search condition C ⁺⁺ is incremented by 1 (step S28).

Next, the microprocessor 6 functioning as the search result output means F determines whether or not the rear node pointer is stored in the node information storage table corresponding to the jth lower node selected in the process of step S25. , that is, it determines whether the j-th lower node selected in step S25 was the last one of the nodes located on one level lower upper node d _i (step S29).

If the determination result in step S29 is false, the subsequent node pointer is stored in the node information storage table corresponding to the jth lower node selected in the process in step S25. In other words, this means that there is another node located one layer lower than the upper node d _i read in the process of step S22, so that the microprocessor 6 functioning as the search result output means F again performs step S24. Then, the process of steps S24 to S29 is repeatedly executed in the same manner as described above based on the updated current value of the index j.

Therefore, finally, when the determination result in step S29 becomes true, all nodes located one hierarchy lower than one upper node d _i stored in the sorting table of FIG. 8 correspond to that node. The attribute information storage table is referred to, and it is determined whether or not the attribute information in the field of the attribute information storage table matches the search condition C ⁺⁺ . Based on the determination result, the attribute corresponding to the lower node is determined. The process of storing the attribute information of the name read from the information storage table in the sorted search result storage table of FIG. 9 in association with the upper node d _i is completed.

That is, when the determination result of step S29 is Yes, C ⁺⁺ attribute information field in the attribute information table of all the subordinate nodes located one level lower in one upper node d _i is a search condition Are detected, all the attribute information of the names stored in the attribute information storage table corresponding to these lower nodes are read, and these names, for example, e ₁ and e _{2 in} FIG. value is to be stored in the sorting already search result storage table shown in FIG. 9 corresponds to the upper node d _i.

Next, the microprocessor 6 that functions as the distribution density calculation function realization means in the search result output means F is the current value of the index j, that is, the final value of the index j, and is one hierarchy lower than the one upper node d _i at that time. C ⁺⁺ , which is the value of the index j indicating the total number of lower nodes located at, and the attribute information of the field in the corresponding attribute information storage table among the nodes located one layer lower than the upper node d _i is the search condition By dividing the value of the index C representing the number of matching lower nodes, the distribution density as already described, that is, the value of [number of lower nodes matching the search condition / total number of lower nodes] is obtained (step S30). , the value of the upper node d _i, more specifically, in correspondence with each of the names that are stored corresponding to the upper node d _i, sorting Sumiken as shown in FIG. 9 Results to be stored as _f 1, _{f 2} in the storage table (step S31).

In this way, with respect to one upper node stored in the sorting table of FIG. 8, the lower node having the attribute information storage table storing the attribute information that matches the search condition C ⁺⁺ among the lower nodes located one hierarchy lower. A process for storing all names that are a kind of attribute information stored in the attribute information storage table of the node in the sorted search result storage table in FIG. 9 in association with the upper node, and corresponding to each name. When [C / j], that is, the process of storing the distribution density in the sorted search result storage table of FIG. 9 is completed, the microprocessor 6 proceeds to the process of step S20 again and sets the updated current value of the index i. Accordingly, the processing from step S20 to step S31 is repeatedly executed in the same manner as described above.

Therefore, when the determination result in step S21 is finally false, all of the upper nodes stored in the sorting table of FIG. 8 match the search condition C ⁺⁺ among the lower nodes located one hierarchy lower. Sorted search result storage table of FIG. 9 in which all the names that are a kind of attribute information stored in the attribute information storage table of the lower node having the attribute information storage table storing the attribute information to be stored are associated with each upper node. And the process of storing the distribution density common to each higher order node in correspondence with each name in the sorted search result storage table of FIG. 9 is completed.

  Finally, the microprocessor 6 functioning as the search result output means F sets the name stored in the sorted search result storage table as shown in FIG. 9 and the distribution density value corresponding to each name as one set. As the data [name, distribution density], the search results are sequentially read out from the top record in the search result storage table as shown in FIG. 9 and output to the outside (step S32).

More specifically, when the search result output device 1 is used alone, the microprocessor 6 functioning as the search result output means F is operated by the sorted search result storage table as shown in FIG. As a set of data [name, distribution density], the name stored in the name and the distribution density value corresponding to each name are read out as a search result from the first record in the search result storage table as shown in FIG. They are displayed on the display 16 in order.
Of course, in the search result output apparatus 1 provided with a printer, the printer may be used as the output destination of the search result instead of the display 16.

  If the search result output apparatus 1 is connected to a network 4 such as the Internet or an in-house LAN and is used from other client machines 5, 5, 5,. The microprocessor 6 functioning as the result output means F reads the search results in order from the first record in the search result storage table as shown in FIG. 9, and sends the search request to the client machine 5 that has transmitted the search request via the network 4. The client machine 5 that has received the request outputs the search result consisting of [name, distribution density] in the same order as described above to the display or printer of the client machine 5.

As described above, in this embodiment, first, the search means B accesses the database 3 stored in the hard disk drive 10 functioning as the storage device 2, and stores attribute information that matches the search condition such as C ^++. After the search of the attribute information storage table is completed and a keyword-corresponding search result storage table as shown in FIG. 6 is generated (steps S1 to S4), the upper node detecting means C accesses the database 3 to store the node information. Referring to the table, the upper node located one level higher than the node corresponding to the attribute information storage table storing the attribute information matching the search condition is detected without duplication and stored in the count-up table as shown in FIG. (Step S5 to Step S10), the counting means D is positioned one hierarchy lower than the detected upper node. The count-up table as shown in FIG. 7 is obtained by counting the number of nodes corresponding to the attribute information storage table storing attribute information that matches the search condition such as C ⁺⁺ among the lower nodes for each upper node. (Step S11 to step S17).
Further, based on the magnitude of the counted number of lower nodes, the sorting means E changes the rank of the upper nodes from the detection order stored in the count-up table as shown in FIG. 7 as shown in FIG. The sorting result is temporarily stored in the sorting table in the descending order (step S18), and the search result output means F stores the lower-level nodes positioned one hierarchy lower than the upper-level nodes in the order of the higher-order nodes sorted in the sorting table. Of the attribute information stored in the attribute information storage table storing attribute information in the field that matches the search condition, the attribute information of the type specified as the output target, that is, in this example, the name and the [search condition And the distribution density indicated by the value of the number of subordinate nodes matching the total number of subordinate nodes] is stored as a sorted search result as shown in FIG. It is temporarily stored in Buru (step S19~ step S31).
Finally, the search result output means F outputs the name and distribution density as the search results in order from the top of the sorted search result storage table, that is, in the order of the upper nodes rearranged by the sorting means E. (Step S32).
Therefore, the contents and number of attribute information output as a search result, that is, in this case, the contents and number of names as search results are immediately output as search results by the well-known search means B and search means B. This is the same as the case of a general search result output device, search result output control program, and search result output method having only known search result output means, but a hierarchical structure of reference attribute information, that is, this implementation Attribute information that stores the attribute information that matches the search condition in the same hierarchy in this hierarchical structure, using the personal status, which is the attribute information that is the reference in the form, as another standard different from the search condition The search results are listed in descending order of the distribution density expressed by the hierarchy having a large number of nodes corresponding to the storage table, that is, the number of lower nodes matching the search condition / the total number of lower nodes. That name will be is output.

Therefore, even when there are a plurality of search results having the same degree of match with the search condition, for example, a plurality of individuals with the same degree of familiarity with the search condition C ⁺⁺ , the output order of the search results is confirmed. In this embodiment, in particular, by checking the value of the distribution density value, the superiority and inferiority of the search results are differentiated, and the user easily obtains the search results that seem more appropriate. It becomes possible.

FIG. 14 is a conceptual diagram showing an example of output of search results from the display 16 and the printer. FIG. 14A shows an example of output when the present embodiment is applied, and FIG. The attribute information storage table of each individual is referred to determine whether the attribute information of each individual matches a field such as a search condition, for example, C ⁺⁺ , and the output target of the attribute information stored in the matching attribute information storage table An example of output in the case where the attribute information of the type designated as, that is, in this case, the personal name is immediately output to the outside is shown.

  In a general search result output device, a search result output control program, and a search result output method having only a known search means B and a known search result output means for immediately outputting a search result by the search means B, the search condition is met. As shown in FIG. 14B, when a plurality of search results having the same degree of match with the search condition are output, the search results when the degrees are the same are output for the reasons described above. The order is in accordance with the registration order of the attribute information storage table and the alphabetical order of the names of each individual, and it is difficult for the user to specify the superiority or inferiority of the search results, but in this embodiment, For example, as shown in FIG. 14A, the superiority or inferiority of the search results can be easily identified by checking the output order of the search results and the size of the distribution density corresponding to the search results. Rukoto can.

The reason why the high distribution density on the directory tree representing the hierarchical structure of the organization is important in the search results is as follows.
First, people who are densely located in the same hierarchy due to the positional relationship on the directory tree have a higher degree of relevance in the organization than those who are discrete, and are considered to belong to the same department or close departments. It is done.
This indicates that this department is also in charge of a specific field, for example, a field such as C ⁺⁺ , or a department where people who are familiar with the field gather.
Therefore, people who belong to departments with high distribution density are those who are more suitable for the purpose, that is, those who are most familiar with the field such as C ⁺⁺ , which is a search condition, and are the central person in the network of people and organizations related to this field. By preferentially outputting these people as search results, the names of people who are particularly familiar with the field are more likely to be output as search results for the first group. This is because it is considered that a plurality of search results having the same degree of match with the search condition are effectively differentiated.

The embodiment described above corresponds to the attribute information storage table that stores attribute information that matches the search condition only for the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means C. The number of nodes is counted by the counting means D, but the attribute information that matches the search condition is stored in two or more hierarchies immediately below the upper node detected by the upper node detecting means C. It is also possible to count the number of nodes corresponding to the attribute information storage table to be counted by the counting means D, and rearrange the rank of the upper nodes by the sorting means E based on the counting result.
In this way, if the range of the directory tree (hierarchy of organizational units) that performs the counting process to obtain the high distribution density is expanded a little, the organizational hierarchy on the directory tree is deeply subdivided, and Even in the situation where nodes matching the search conditions are dispersed in each department at the end department, the count value by the counting means D, that is, the distribution density is positively given a clear difference and sorting Sorting by means E can be performed appropriately.
It is also possible to variably specify the range of the directory tree in which the counting process for obtaining the high distribution density is performed by the parameter setting of the search result output device 1 or the client 5.

  The search means B, the upper node detection means C, the counting means D, the sorting means E, and the search result output means F are not necessarily configured by a single microprocessor, and an independent microprocessor is used for each means. May be. In that case, it is necessary to transfer intermediate data obtained by each means between the microprocessors. For example, a part of the storage area of the RAM 9 is used as a shared memory, and the keyword-corresponding search result is stored in this shared memory. If the storage table, count-up table, sorting table, sorted search result storage table, etc. are temporarily stored so that reading and writing from the microprocessor for each means is allowed, data transfer between the microprocessors is easy. is there.

  A part or all of the embodiment disclosed above can be appropriately expressed by the description shown in the following supplementary notes, but the form for carrying out the invention and the technical idea of the invention are not limited to these. Absent.

  The search result output control program may be stored in a non-temporary recording medium. As the non-temporary recording medium, for example, a DVD, CD, flash memory or the like can be used. When the search result output control program is stored in a non-temporary recording medium, the search result output control program is read by the computer from the non-temporary recording medium and executed by the microprocessor.

[Appendix 1]
A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes The attribute information storage table storing the attribute information is connected to a storage device previously stored as a database, the attribute information storage table in the database is searched based on the search condition input from the interface, and the attribute information matches the search condition A search result output device for outputting, as a search result, attribute information of a type designated as an output target among attribute information stored in the attribute information storage table.
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Upper node detecting means for detecting nodes without duplication referring to the node information storage table;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means to perform,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means;
Search the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node according to the order of the upper nodes rearranged by the sorting means. A search result output device comprising search result output means for outputting a result.

[Appendix 2]
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located in a plurality of lower layers immediately below the upper node detected by the upper node detecting unit by the counting unit The search result output device according to supplementary note 1, wherein: is counted for each upper node.

[Appendix 3]
The reference attribute information is the position that each individual occupies in the organizational structure of the functional group, and the attribute information storage table stores at least attribute fields for identifying each individual and attribute information for identifying each individual, At least attribute information of a field is input to the interface as the search condition, and the search result output means outputs attribute information for identifying each individual as the specified type of attribute information that is the search result. The search result output device according to appendix 1 or appendix 2.

[Appendix 4]
A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes A computer connected to a storage device that stores the attribute information storage table storing the information as a database in advance,
Search means for searching an attribute information storage table storing attribute information matching the search condition from the attribute information storage table in the database based on a search condition input from an interface;
When the search by the search means is completed, an upper node located one level higher than the node corresponding to the attribute information storage table storing the attribute information matching the search condition is duplicated with reference to the node information storage table Upper node detecting means for detecting without,
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means; and
Of the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node in the order of the upper nodes rearranged by the sorting means A search result output control program that functions as search result output means for outputting attribute information of a type specified as an output target as a search result.

[Appendix 5]
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located in a plurality of lower layers immediately below the upper node detected by the upper node detecting unit by the counting unit The search result output control program according to appendix 4, characterized in that the number is counted for each upper node.

[Appendix 6]
A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes The attribute information storage table storing the attribute information is stored in advance as a database, the attribute information storage table in the database is searched based on a given search condition, and the attribute information that matches the search condition is stored. A search result output method for outputting, as a search result, attribute information of a type specified as an output target among attribute information stored in a storage table,
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Detecting nodes with reference to the node information storage table without duplication;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the detected upper node is counted for each upper node,
Rearranging the rank of the upper node in descending order based on the number of counted lower nodes,
The attribute information stored in the attribute information storage table storing the attribute information that matches the search condition among the lower nodes positioned one hierarchy lower than each upper node in accordance with the rearranged upper node order is output as the search result. A search result output method characterized by that.

[Appendix 7]
Attribute information that matches the search condition among a plurality of lower nodes located in a plurality of lower layers immediately below the node detected as an upper node of the node corresponding to the attribute information storage table that stores attribute information that matches the search condition The number of nodes corresponding to the attribute information storage table to be stored is counted for each upper node, and the order of the upper nodes is rearranged in descending order based on the number of counted lower nodes. The search result output method according to appendix 6.

[Appendix 8]
The reference attribute information is the position that each individual occupies in the organizational structure of the functional group, and the attribute information storage table stores at least attribute fields for identifying each individual and attribute information for identifying each individual, The search result according to appendix 6 or appendix 7, wherein at least attribute information of a field is input as a search condition, and attribute information for specifying each individual is output as attribute information of a specified type as a search result output method.

  The present invention can be used for various search processes using a database having a directory tree, particularly for a search system with a high possibility of obtaining a plurality of search results having the same degree of match with a search condition.

1 Search result output device (computer)
2 storage device 3 database 4 network 5 client machine 6 microprocessor (search means, upper node detection means, counting means, sorting means, search result output means)
7 ROM
8 Nonvolatile memory 9 RAM
10 Hard disk drive (storage device)
11 Communication interface (interface)
12 External drive 13 Output circuit 14 Keyboard (interface)
15 mouse 16 display A interface B search means C upper node detection means D counting means E sorting means F search result output means P1 to P46 nodes

Claims (8)

A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes The attribute information storage table storing the attribute information is connected to a storage device previously stored as a database, the attribute information storage table in the database is searched based on the search condition input from the interface, and the attribute information matches the search condition A search result output device for outputting, as a search result, attribute information of a type designated as an output target among attribute information stored in the attribute information storage table.
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Upper node detecting means for detecting nodes without duplication referring to the node information storage table;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means to perform,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means;
Search the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node according to the order of the upper nodes rearranged by the sorting means. A search result output device comprising search result output means for outputting a result.   The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located in a plurality of lower layers immediately below the upper node detected by the upper node detecting unit by the counting unit The search result output device according to claim 1, wherein: is counted for each upper node.   The reference attribute information is the position that each individual occupies in the organizational structure of the functional group, and the attribute information storage table stores at least attribute fields for identifying each individual and attribute information for identifying each individual, At least attribute information of a field is input to the interface as the search condition, and the search result output means outputs attribute information for identifying each individual as the specified type of attribute information that is the search result. The search result output device according to claim 1 or 2. A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes A computer connected to a storage device that stores the attribute information storage table storing the information as a database in advance,
Search means for searching an attribute information storage table storing attribute information matching the search condition from the attribute information storage table in the database based on a search condition input from an interface;
When the search by the search means is completed, an upper node located one level higher than the node corresponding to the attribute information storage table storing the attribute information matching the search condition is duplicated with reference to the node information storage table Upper node detecting means for detecting without,
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the upper node detected by the upper node detecting means is counted for each upper node. Counting means,
Sorting means for rearranging the rank of the higher order nodes in descending order based on the number of lower order nodes counted by the counting means; and
Of the attribute information stored in the attribute information storage table storing the attribute information matching the search condition among the lower nodes located one hierarchy lower than each upper node in the order of the upper nodes rearranged by the sorting means A search result output control program that functions as search result output means for outputting attribute information of a type specified as an output target as a search result.   The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located in a plurality of lower layers immediately below the upper node detected by the upper node detecting unit by the counting unit 5. The search result output control program according to claim 4, wherein: is counted for each upper node. A node information storage table storing node specifying information for specifying each node of a directory tree representing a hierarchical structure of reference attribute information for each node, and attribute information other than the reference attribute information corresponding to each of the nodes The attribute information storage table storing the attribute information is stored in advance as a database, the attribute information storage table in the database is searched based on a given search condition, and the attribute information that matches the search condition is stored. A search result output method for outputting, as a search result, attribute information of a type specified as an output target among attribute information stored in a storage table,
When the search of the attribute information storage table that stores the attribute information that matches the search condition is completed, the upper level that is located one level higher than the node corresponding to the attribute information storage table that stores the attribute information that matches the search condition Detecting nodes with reference to the node information storage table without duplication;
The number of nodes corresponding to the attribute information storage table that stores the attribute information that matches the search condition among the lower nodes located one hierarchy lower than the detected upper node is counted for each upper node,
Rearranging the rank of the upper node in descending order based on the number of counted lower nodes,
The attribute information stored in the attribute information storage table storing the attribute information that matches the search condition among the lower nodes positioned one hierarchy lower than each upper node in accordance with the rearranged upper node order is output as the search result. A search result output method characterized by that.   Attribute information that matches the search condition among a plurality of lower nodes located in a plurality of lower layers immediately below the node detected as an upper node of the node corresponding to the attribute information storage table that stores attribute information that matches the search condition The number of nodes corresponding to the attribute information storage table to be stored is counted for each upper node, and the order of the upper nodes is rearranged in descending order based on the number of counted lower nodes. The search result output method according to claim 6.   The reference attribute information is the position that each individual occupies in the organizational structure of the functional group, and the attribute information storage table stores at least attribute fields for identifying each individual and attribute information for identifying each individual, 8. The attribute information for specifying each individual is output as attribute information of a specified type as a search result, and at least attribute information of a field is input as a search condition. Search result output method.

Images