JP2014194608A - Retrieval system, retrieval method and retrieval program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a retrieval system capable of performing a condition-designated narrowing-down retrieval on a specific attribute that text has.SOLUTION: A retrieval system comprises: a display section 203; an input section 201; an input-enabled candidate list 404; a substance DB 402; an input-enabled attribute name display processing section 202 for displaying on the display section 203 input-enabled attribute information relative to a retrieval character string; a retrieval key creation section 301 for creating a retrieval key including attribute information on the basis of attribute information and the retrieval character string inputted from the input section 201 based on the input-enabled attribute information displayed on the display section 203; and a substance data acquisition section in which the substance DB 402 is retrieved on the basis of retrieval information of multi-dimensional data based on the retrieval key and a retrieval result is displayed on the display section 203.

Description

  The present invention relates to a search system, and in particular, text described in a free format for multi-dimensional data having both a text described in a free format and a value in which a format such as a category value or a date is determined. The present invention relates to a search method that combines a text search for a sentence such as, and an attribute search for attributes such as category values and dates other than text.

  In multidimensional data, a search is often performed to find data according to the purpose from the data set. There are two methods of search, text search and attribute search.

  The attribute search is a method often used when data is extracted from a table of a relational database (hereinafter referred to as RDB), and is a method of extracting data that satisfies a condition by assigning a condition to the attribute.

  The text search is a method of extracting data including a search word input by a user in the text, like a search engine such as Google (registered trademark).

  Furthermore, a method of performing a text search by narrowing down the range of search target data by attribute has been proposed. For example, Japanese Patent Application Laid-Open No. 2004-289848 (Patent Document 1) compares an attribute list having a search word input by the user in advance with an attribute list that is not included in the attribute search or attribute list. In this case, a method for automatically searching for text search is shown.

JP 2004-289848 A

  In the attribute search, there is a problem that the user needs to know in advance what kind of attribute exists in the multidimensional data. As a general solution to this problem, there is a method of displaying an attribute name in advance in a user interface (hereinafter referred to as UI). In this method, whenever a user changes an attribute name or adds an attribute, It takes time and effort to change the UI.

  In the text search, the user can input only the search word and extract data including the search word input from the attribute and the text, so that the search can be performed without knowing the attribute name. Also, there is no need to change the UI.

  However, it cannot be narrowed down by a data set having an attribute having a specific property, and the number of data items obtained as a search result increases, so that it is difficult to find target data.

  In Patent Document 1, since search is automatically performed by determining whether a search word is a text search or an attribute search, a text search can be performed by narrowing down the range of search target data by attribute, and which attribute includes which value. There is no need to know in advance.

  However, in Patent Document 1, it is not possible to perform a narrow search by specifying a condition for a specific attribute. When performing a text search by narrowing down search target data with a specific attribute, the user cannot perform a search under intended conditions.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a search system that can perform a narrow search that specifies a condition for a specific attribute of a text.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  That is, a typical overview is a search system that performs a search process for multidimensional data including sentences described in a free format, and displays information related to the search process to the user. A search character string from the user based on display information on the display unit and an input unit for inputting attribute information for the search character string, an inputable candidate list in which attribute information that can be input for the search character string is stored, An inputable attribute name that causes the display unit to display attribute information that can be entered for the search character string based on the entity database that stores information about multidimensional data, the search character string input from the input unit, and the inputable candidate list Based on attribute information and a search character string input from a display processing unit and an input unit based on inputable attribute information displayed on the display unit, the attribute A search key creation unit that creates a search key including information, and an entity data acquisition unit that searches the entity database based on the search information of the multidimensional data based on the search key and displays the search result on the display unit. It is provided.

  In addition, a display unit that displays information related to search processing for the user, an input unit that inputs a search character string from the user based on the display information of the display unit, and attribute information for the search character string, and a search character string Search for multidimensional data including sentences written in free format, including an inputable candidate list in which attribute information that can be input is stored and an entity database in which information on multidimensional data is stored A search method in a search system that performs processing, wherein an inputable attribute name display processing unit obtains inputable attribute information for the search character string based on a search character string and an inputable candidate list input from the input unit. The attribute information input from the input unit based on the inputable attribute information displayed on the display unit at the step of displaying on the display unit and the search key creating unit. And a step of creating a search key including attribute information based on the search character string, and the entity data acquisition unit searches the entity database based on the search information of the multidimensional data based on the search key, and obtains the search result. And displaying on the display unit.

  In addition, a display unit that displays information related to search processing for the user, an input unit that inputs a search character string from the user based on the display information of the display unit, and attribute information for the search character string, and a search character string Multidimensional including sentences described in free format using a computer system comprising an inputable candidate list storing inputable attribute information for and a substance database storing information on multidimensional data A search program for performing a search process for data, wherein a computer system is configured to display attribute information that can be input to a search character string based on a search character string and an inputable candidate list input from the input unit. Inputable attribute name display processing section to be displayed, and attribute information and verification input from the input section based on the inputable attribute information displayed on the display section. A search key creation unit that creates a search key including attribute information based on a character string, and a search of a substance database based on multi-dimensional data search information based on the search key, and a search result is displayed on a display unit. It functions as an entity data acquisition unit.

  The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows.

  In other words, in a search for multidimensional data including sentences described in a free format, the user can perform a text search by narrowing down the range of the search target data by attribute search. It becomes easy to extract data.

It is a block diagram which shows the structure of the search system which concerns on one embodiment of this invention. It is a flowchart which shows the whole operation | movement of the search system which concerns on one embodiment of this invention. It is a flowchart which shows the process sequence which the input part in FIG.2 S101 of the search system which concerns on one embodiment of this invention receives an input from a user, and displays the attribute name which can be input in an inputable attribute name display process part. . It is a figure which shows the example of a screen which receives the search conditions of the search system which concerns on one embodiment of this invention. It is a figure which shows the schema of the input possibility candidate list | wrist of the search system which concerns on one embodiment of this invention, and a data storage example. It is a figure which shows the example of a screen of the input possible candidate of the search system which concerns on one embodiment of this invention. It is a figure which shows the example of a screen which displays the attribute name used as the next input possibility candidate after one attribute of the search system which concerns on one embodiment of this invention is input. It is a flowchart which shows the process sequence in which the search key preparation part in FIG.2 S102 of the search system which concerns on one embodiment of this invention produces a search key. It is a figure which shows the schema of the attribute name list of the search system which concerns on one embodiment of this invention, and a data storage example. It is a flowchart which shows the process sequence in which the index search part in FIG.2 S103 of the search system which concerns on one embodiment of this invention acquires data ID by search of a search key. It is a figure which shows the schema of the index DB of the search system which concerns on one embodiment of this invention, and a data storage example. It is a figure which shows the schema of the corresponding list of the search system which concerns on one embodiment of this invention, and a data storage example. 2 is a processing procedure in which the entity data ID list creation unit in step S104 of FIG. 2 of the search system according to the embodiment of the present invention creates an entity data ID list for acquiring the entity data by filtering the data ID. It is a flowchart which shows. It is a figure which shows the data ID merge example of the corresponding | compatible list of the search system which concerns on one embodiment of this invention. It is a figure which shows the creation example of the entity data ID list of the search system which concerns on one embodiment of this invention. It is a flowchart which shows the process sequence in which the entity data acquisition part in step S105 of FIG. 2 of the search system which concerns on one embodiment of this invention acquires entity data. It is a figure which shows the schema and data storage example of entity DB of the search system which concern on one embodiment of this invention. It is a figure which shows the example of the search result display screen of the search system which concerns on one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

<Configuration of search system>
A configuration of a search system according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a configuration diagram showing a configuration of a search system according to an embodiment of the present invention.

  As shown in FIG. 1, a search management subsystem 200, a search subsystem 300, and a search target database (DB) group having a computer system 101 and its user 100 constituting the search system and arranged on the computer system 101. The mutual relationship of 400 is shown typically.

  The computer system 101 is composed of one or more computers having a calculation processing capability such as a central processing unit.

  As a specific example of the computer system 101, each processing function can be realized by executing a program in a computer such as a PC or a server having elements such as a processor, a memory, a disk, a bus, a communication interface unit, an input device, and an output device.

  For example, the processing of each unit is realized by a processor loading and executing a search program or other control program from a disk or the like to a memory.

  The search management subsystem 200 includes an input unit 201, an inputable attribute name display processing unit 202, and a display unit 203.

  The input unit 201 inputs a search character string from the user 100 and attribute information such as an attribute name for the search character string, and the display unit 203 displays information related to the search process to the user 100.

  The input unit 201 and the display unit 203 can be configured by, for example, control processing of a keyboard, mouse, and other input devices by a computer, control processing of an output device such as a display, and the like.

  The search management subsystem 200 receives a character string as a search character string expressing a search condition from the user 100 via the input unit 201 and passes it to the search subsystem 300.

  When the user 100 inputs a search word, the inputable attribute name display processing unit 202 causes the display unit 203 to display an attribute name having the search word as a value.

  The search condition is expressed by a search word or a combination of a search word and an attribute name. The search character string can also include a plurality of search conditions. Further, the search subsystem 300 receives search results extracted from the search target DB group 400 and displays them on the display unit 203.

  The search subsystem 300 is a system that extracts data from the search target DB group 400 according to a search character string.

  The search subsystem 300 includes a search key creation unit 301, an index search unit 302, an entity data ID list creation unit 303, and an entity data acquisition unit 304.

  The search key creation unit 301 has a function of creating a character string serving as a search key that combines a search word and an attribute bit according to a search condition.

  The index search unit 302 has a function of extracting the data ID corresponding to the search key from the index DB 401.

  The entity data ID list creation unit 303 has a function of filtering IDs according to search conditions and creating an ID list of data to be extracted as a search result.

  The entity data acquisition unit 304 has a function of extracting data from the entity DB 402 based on the ID list.

  The search target DB group 400 stores data to be searched, and includes an index DB 401, an entity DB 402, an attribute name list 403, and an inputable candidate list 404.

  The index DB 401 stores an index of a data ID corresponding to the search key.

  The entity data DB 402 stores data that is entity data in which the text and attributes associated with the data ID are combined into one.

  The attribute name list 403 stores a list of attribute names of the entity data.

  The inputable candidate list 404 stores a list of candidate attributes that can be input for the search string value.

<Overall operation of search system>
Next, the overall operation of the search system according to the embodiment of the present invention will be described with reference to FIG. FIG. 2 is a flowchart showing the entire operation of the search system according to the embodiment of the present invention, and shows the search procedure of the entire system when the search system performs data search.

  In the following description, blog search is described as an example.

  First, the search management subsystem 200 receives a search character string from the user via the input unit 201 (step S101).

  A plurality of search character strings can be input by separating a search word or a combination of a search word and one or more attribute names with a delimiter (for example, a blank character). Here, a search word or a combination of a search word and an attribute name is called a search condition.

  The search word and attribute name are separated from the search condition delimiter (for example, a colon). When multiple attribute names are specified, the delimiter is different from the delimiter (for example, a comma). Etc.).

  The received search character string is transferred to the search key creation unit 301 of the search subsystem 300, and a search key is created (step S102). The search key is a character string obtained by concatenating a search word and an attribute bit indicating which attribute the search word is specified for.

  Thereafter, the search key is transferred to the index search unit 302. The index search unit 302 searches the index DB 401 using the search key as a key, and acquires a list of data IDs that match the search key condition (step S103). .

  One list of data IDs is acquired for one search key. Thereafter, the list of search keys and data IDs is sent to the entity data ID list creation unit 303, and an entity data ID list that is an entity data ID list that meets all the conditions is created in accordance with the specified conditions (step S104). ).

  The entity data ID list is delivered to the entity data acquisition unit 304, and the entity DB 402 is searched using the data ID included in the entity data ID list as a key to acquire the entity data (step S105).

  Finally, the acquired entity data is transferred to the display unit 203, the search result is displayed (step S106), and the process ends.

<Detailed operation of search system>
Next, a detailed operation of the search system according to the embodiment of the present invention will be described with reference to FIGS.

  3 to 18 are explanatory diagrams for explaining the detailed operation of the search system according to the embodiment of the present invention, and show details of steps S101 to S105 shown in FIG.

  FIG. 3 is a flowchart showing a processing procedure in which the input unit 201 in step S101 in FIG. 2 receives input from the user and the inputable attribute name display processing unit 202 displays inputable attribute names, and FIG. 4 is a screen for receiving search conditions. It is a figure which shows an example.

  FIG. 5 is a diagram showing a schema 601 and a data storage example 602 of the input-possible candidate list 404, FIG. 6 is a diagram showing an example of an input-possible candidate screen, and FIG. 7 shows the next input-possible candidate after one attribute is input. It is a figure which shows the example of a screen which displays the attribute name which becomes.

  FIG. 8 is a flowchart showing a processing procedure for creating a search key by the search key creation unit 301 in step S102 of FIG. 2, and FIG. 9 is a diagram showing an attribute name list schema and a data storage example.

  FIG. 10 is a flowchart showing a processing procedure in which the index search unit 302 obtains a data ID by searching for a search key in step S103 of FIG. 2, FIG. 11 shows a schema of the index DB 401 and a data storage example, and FIG. It is a figure which shows the schema of a list, and a data storage example.

  FIG. 13 is a flowchart showing a processing procedure in which the entity data ID list creation unit 303 in step S104 of FIG. 2 creates an entity data ID list for acquiring the entity data by filtering the data ID, and FIG. FIG. 15 is a diagram showing an example of data ID merging of lists, and FIG. 15 is a diagram showing an example of creating an entity data ID list.

  16 is a flowchart showing a processing procedure in which the entity data acquisition unit 304 acquires entity data in step S105 of FIG. 2, FIG. 17 is a diagram showing an example of the schema and data storage of the entity DB 402, and FIG. 18 is an example of a search result display screen. FIG.

[Processing Procedure of Step S101 in FIG. 2]
First, as shown in FIG. 3, the input unit 201 accepts input from the user in step S101 of FIG. 2 and displays the inputable attribute name in the inputable attribute name display processing unit 202. An input of a search character string is received from the user (step S201).

  FIG. 4 shows an example of a UI that accepts user input at this time.

  As shown in FIG. 4, the UI has a text field 501 for accepting input and a search execution button 502 for pressing a search at the time of execution.

  Next, it is determined whether or not a delimiter between the search word and the attribute name is input (step S202). Here, as an example, the delimiter between the search word and the attribute name will be described as “: (colon)”, but any character can be used as long as the delimiter can be determined.

  When a colon as a delimiter is input in step S202, the input search word is acquired, and the input possibility candidate list 404 is searched to acquire the attribute bits of the search word (step S203).

  If a colon as a delimiter is not input in step S202, the process proceeds to input end determination (step S208).

  FIG. 5 shows a schema 601 and a data storage example 602 of the input possibility candidate list 404. For a text portion such as a blog body, the text is divided into words by morphological analysis, and inputable candidates are stored for the word. An inputable candidate is an attribute name having the word as a value.

  Then, the inputable attribute name display processing unit 202 displays the inputable candidates obtained by searching the inputable candidate list 404 in the text field (step S204).

  In the case of the example shown in FIG. 5, when a search word “travel” is input and a colon is input, “category” and “blog text” are acquired as input possible candidates. The acquired input possible candidates are displayed as a list.

  FIG. 6 shows an example of a screen for inputable candidates. In the example shown in FIG. 6, “category” and “blog text” are displayed.

  When one attribute name is selected from the displayed list of input candidates, the selected attribute name is input to the text field (step S205).

  Then, it is determined whether or not a comma as an attribute name delimiter is input after the attribute name is input (step S206). If a comma as an attribute name delimiter is input in step S206, the process returns to the process of step S204 for displaying inputable candidates, and the attribute name is displayed.

  At this time, attribute names that have already been input are excluded from the inputable candidates to be displayed. A display example of the input-capable candidates to be displayed is shown in FIG. In the example illustrated in FIG. 7, for example, the attribute name that has already been input is “blog body”, and only “category” is displayed.

  The attribute name input process (steps S204 to S205) can be repeated for the input of one search word, and a plurality of attribute names can be input for one search word. .

  If a comma as an attribute name delimiter is not input after the attribute name is input in step S206, it is determined whether or not a blank character as a search condition delimiter is input (step S207). If a blank character is input in step S207, the process returns to step S201 to accept another search condition.

  If no blank character is input in step S207, the input end determination is performed (step S208).

  The case where it is determined that the input is finished is, for example, when the search execution button 502 indicated by the UI shown in FIG. 4 is pressed.

  If it is determined in step S208 that the input has been completed, the search character string input in the text field is sent to the search key creation unit (step S209). If it is not determined in step S208 that the input has been completed, the process returns to step S201. .

[Processing Procedure of Step S102 of FIG. 2]
The processing procedure for creating the search key by the search key creation unit 301 in step S102 in FIG. 2 is as follows. First, as shown in FIG. And the search character string is divided into search conditions by a delimiter (for example, a blank character) (step S301).

  Next, in order to create a search key from the search condition, it is determined whether there is a search condition for which no search key has been created (step S302).

  If there is a search condition for which no search key has been created in step S302, the search condition is acquired to determine whether an attribute name is specified (step S303).

  If an attribute name is specified in step S303, the search condition is divided into a search word and an attribute name by a delimiter (for example, a colon) (step S304), and the attribute name list 403 is read (step S305).

  Thereafter, the attribute name list 403 is compared with the input attribute name to create an attribute bit (step S306).

  FIG. 9 shows a schema 701 and a storage example 702 of the attribute name list 403.

  The attribute name list 403 stores attribute names of data. When there are a plurality of input attribute names, they are divided by a delimiter (for example, a comma) and each is compared with an attribute list.

  For example, when the attribute name list 403 is {category, creator, creation date, update date, blog body text} and the input attribute name is “creator”, “creator” is the second from the left in the attribute name list. Therefore, an attribute bit of “01000” is obtained in which the bit is 1 and the remaining bits are 0. In the case of “category, blog text”, the attribute bit “10001” is obtained.

  When the attribute bit is obtained, the search bit is created by connecting the attribute bit and the input search word with a delimiter (for example, a colon) (step S308).

  For example, when “travel: blog text” is a search condition, the search key is “00001: travel”.

  If the attribute name is not specified in step S303 and only the search word is specified, all the attributes and text are to be searched, and attribute bits of all patterns are created (step S307).

  If there are five attribute name lists, all patterns of five digits except “00000” indicating that all values are not included, that is, “00001” “00010”... “11110” “11111” Are created as attribute bits.

  When the attribute bits are created, search keys for all patterns except 00000 for which no attribute is selected are created (step S308).

  For example, if “Okinawa” is the search condition, 31 search keys of “00001: Okinawa”, “00010: Okinawa”... “11110: Okinawa”, “11111: Okinawa” are created.

  When the search key is created, the created search key is added to the search key list (step S309).

  The search key list is a list for holding the created search keys, and is held in the format of {10000: Travel, 00001: Okinawa, 00010: Okinawa, ..., 11110: Okinawa, 11111: Okinawa}.

  If a search key is added in step S309, the process returns to the determination in step S302 as to whether there is a search condition for which no search key has been created.

  If there is no search condition for which no search key has been created in step S302 and search keys have been created for all search conditions, the search key list is sent to the index search unit 302 (step S310).

[Processing Procedure of Step S103 in FIG. 2]
In addition, as shown in FIG. 10, the index search unit 302 first searches the search key creation unit 301 from the search key creation unit 301 in order to obtain the data ID by searching the search key in step S103 of FIG. A key list is acquired (step S401).

  When the search key list is acquired, a key matching the search key is searched from the index DB 401, and a list of data IDs associated with the key is acquired (step S402).

  FIG. 11 shows a schema 801 of the index DB 401 and a data storage example 802.

  The index DB 401 stores data in which a character string obtained by combining attribute bits and attribute values stored in the attributes with a delimiter (for example, a colon) is a key, and a list of data IDs that meet the key conditions is a value. Has been.

  The index DB 401 stores an index for text in addition to an index for attributes.

  An attribute bit is a character string that indicates which attribute value the attribute value is included in. The number of bits is the same as the number of columns in the data. If the value is included, 1 is included. Is represented by zero.

  An index for text is created for each word divided by morphological analysis of the text.

  As an example, there are {category, creator, creation date, update date, blog text} in the data, and {travel, Tanaka, 20120925, 201202025, and Okinawa, respectively, went to play. }, The attribute bits for “travel” are “10001”, the attribute bits for “Tanaka” are “01000”, and the attribute bits for “20120925” are “00110”.

  In addition, the text “I went to Okinawa to play”, which is the blog text, is divided into the words “Okinawa”, “Play” and “I went”, and an attribute bit of “00001” is attached.

  At this time, the words may be indexed with respect to the converted words by performing name identification such as making a verb verb. The data ID is a number assigned to uniquely identify data, and corresponds to a primary key in the RDB.

  When the list of data IDs is acquired, a correspondence list between the search key and the list of data IDs is created (step S403). FIG. 12 shows a correspondence list schema 901 and a part 902 of a storage example.

  After the correspondence list is created, the correspondence list is sent to the entity data ID list creation unit 303 (step S404).

[Processing Procedure of Step S104 in FIG. 2]
Further, the processing procedure for the entity data ID list creation unit 303 in step S104 of FIG. 2 to create the entity data ID list for acquiring the entity data by filtering the data ID is as shown in FIG. First, the entity data ID list creation unit 303 acquires a correspondence list from the search key creation unit 301 (step S501).

  After acquiring the correspondence list, the list of data IDs is merged for each search condition (step S502). The correspondence list is created by the correspondence between the search key and the data ID. However, since a plurality of search keys may be created from one search condition, it is necessary to merge the data IDs.

  Since the search condition always has one search word, the data IDs associated with the search keys matching the search word are collected in one list.

  An example of merging data ID lists is shown in FIG. The correspondence list 1001 has search keys and data IDs corresponding to the search conditions “travel: category” and “Okinawa”.

  Since “Okinawa” has 31 search keys, the search keys are merged into a list of one data ID. After merging, a list 1002 of data IDs corresponding to two search words is created.

  After merging the data IDs of the correspondence list, an entity data ID list of the entity data to be acquired is created from the merged correspondence list created using the search word as a key (step S503).

  As an example, when all the input search conditions are AND conditions this time, a product set of data ID lists possessed by each search word is obtained, and the product set is used as an entity data ID list.

  The search condition at the time of input can be set as a delimiter (eg, &, |, etc.), and the entity data ID list can be obtained using a logical expression (AND, OR) corresponding to each delimiter as a condition.

  An example of creation of the entity data ID list is shown in FIG. When the entity data ID list 1102 is obtained by ANDing all the search conditions in the merged correspondence list 1101 of FIG. It is obtained as an ID list.

  Finally, the entity data ID list is sent to the entity data acquisition unit 304 (step S504).

[Processing Procedure of Step S105 in FIG. 2]
Further, the processing procedure in which the entity data acquisition unit 304 acquires the entity data in step S105 of FIG. 2 is as follows. First, the entity data acquisition unit 304 receives the entity data ID from the entity data ID list creation unit 303 as shown in FIG. A list is acquired (step S601).

  Then, the entity data of the data ID stored in the entity data ID list is acquired from the entity DB 402 (step S602).

  FIG. 17 shows a schema 1201 of the entity DB 402 and a data storage example 1202.

  When the entity data is acquired from the entity DB 402, the acquired entity data is sent to the display unit 203 (step S603).

  The display unit 203 receives the entity data from the entity data acquisition unit 304 and displays the search result on the search result display screen (step S106 in FIG. 2). An example of a search result display screen is shown in FIG.

  As shown in FIG. 18, the search result display screen displays a text field 1301 for displaying the input search condition, a table 1302 for displaying a list of search results, and one attribute selected from the search result list. Has a text field 1303.

  In the text field 1301 for displaying the input search condition, the input search condition is displayed.

  If the search condition specifies an attribute name, it is displayed in the format of “attribute name = search word”. If the search condition does not specify an attribute name, the format is “input search word search word”. Is displayed.

  The search result list (1302) displays a list of entity data obtained as search results in a table format.

  In the search result list, all the attributes of the entity data may be displayed as table columns, or only a part of the attributes may be selected by the person who designs the UI.

  In the search result details (1303), all attributes and text of the entity data corresponding to one item selected from the search result list are displayed.

  Through the above processing, in the search for multidimensional data including sentences described in a free format, it becomes possible for the user to perform a text search while narrowing the range of the search target data by attribute search, Extraction of data intended by the user is facilitated.

  In addition, the present invention can assist the work in the case where data retrieval is necessary in a business in which data having both sentences and attributes such as text described in a free format is accumulated.

  For example, in a help desk operation, an operator can search a past inquiry history in response to an inquiry from a user, and answer the user with reference to the search result.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  DESCRIPTION OF SYMBOLS 100 ... User, 101 ... Computer system, 200 ... Search management subsystem, 201 ... Input part, 202 ... Inputable attribute name display processing part, 203 ... Display part, 300 ... Search subsystem, 301 ... Search key creation part, 302 ... index search unit, 303 ... entity data ID list creation unit, 304 ... entity data acquisition unit, 400 ... search target DB group, 401 ... index DB, 402 ... entity DB, 403 ... attribute name list, 404 ... input possible candidate list 501 ... Text field, 502 ... Search execution button.

Claims (5)

A search system that performs search processing for multidimensional data including sentences described in a free format,
A display for displaying information related to the search process to the user;
An input unit for inputting a search character string from the user based on display information of the display unit and attribute information for the search character string;
An inputable candidate list in which attribute information that can be input for the search character string is stored;
An entity database in which information about the multidimensional data is stored;
Based on the search character string input from the input unit and the input possible candidate list, an inputable attribute name display processing unit that displays attribute information that can be input to the search character string on the display unit;
A search key creating unit for creating a search key including the attribute information based on the attribute information and the search character string input from the input unit based on the inputable attribute information displayed on the display unit;
Based on the search information of the multi-dimensional data based on the search key, the entity database is searched, and a search result is displayed on the display unit;
Search system with The search system according to claim 1,
The search system, wherein the search key creation unit creates a search key including the attribute information based on a plurality of attribute information input from the input unit based on the inputable attribute information and the search character string. The search system according to claim 1,
The entity database stores entity data in which text and attributes of the multidimensional data associated with a data ID are combined into one,
An index database storing an index for the attribute information and an index for the text of the search string;
An index search unit that searches the index database based on the search key and obtains a list of the data IDs that match the search key condition;
An entity data ID list creation unit that creates an entity data ID list of search information of the multidimensional data that applies to all conditions based on the conditions specified in the list of data IDs and the attribute information;
Search system with A display unit for displaying information related to search processing for the user, an input unit for inputting a search character string from the user based on the display information of the display unit and attribute information for the search character string, and the search The multidimensional data including a sentence described in a free format, comprising: an inputable candidate list storing inputable attribute information for a character string; and an entity database storing information about the multidimensional data. As a search method in a search system for performing a search process,
In the inputable attribute name display processing unit, based on the search character string input from the input unit and the inputable candidate list, displayable attribute information for the search character string on the display unit;
The search key creation unit creates a search key including the attribute information based on the attribute information input from the input unit based on the inputable attribute information displayed on the display unit and the search character string. Steps,
In the entity data acquisition unit, searching the entity database based on the search information of the multidimensional data based on the search key, and displaying the search result on the display unit;
A search method. A display unit for displaying information related to search processing for the user, an input unit for inputting a search character string from the user based on the display information of the display unit and attribute information for the search character string, and the search Includes sentences written in free format using a computer system having an inputable candidate list storing inputable attribute information for a character string and an entity database storing information on multidimensional data A search program for performing a search process on the multidimensional data,
The computer system is
Based on the search character string input from the input unit and the input possible candidate list, an inputable attribute name display processing unit that displays attribute information that can be input to the search character string on the display unit;
A search key creating unit for creating a search key including the attribute information based on the attribute information and the search character string input from the input unit based on the inputable attribute information displayed on the display unit;
Based on the search information of the multidimensional data based on the search key, search the entity database, and function as an entity data acquisition unit that displays the search result on the display unit,
Search program.

Images