JP2013218380A - Database inquiry sentence analysis device, database inquiry sentence analysis method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a database inquiry sentence analysis device in which a database inquiry sentence having accuracy and high reusability can be efficiently created from an existing database inquiry sentence.SOLUTION: A database inquiry sentence analysis device includes: a load factor detection part 11 for inputting a database inquiry sentence, and detecting, as a load factor imposed on a user when reusing a database inquiry sentence, at least one of a description of predetermined form and a condition formula whose complexity is equal to or more than a prescribed level; and a change information generation part 12 for generating change information for changing the database inquiry sentence to a database inquiry sentence from which a load factor is removed.

Description

  The present invention relates to a database query statement analysis apparatus used for reusing a database query statement.

  In recent years, the amount of information supplied from individuals, private companies, public institutions, and the like has become enormous and updated frequently. As a general mechanism for selectively extracting and using necessary information from such information, a relational database (hereinafter referred to as RDB) or a key-value store suitable for large-scale data. (Hereinafter referred to as KVS). KVS can be handled by a simple search, but an RDB, XML database, object-oriented database, or the like is required to support a search using more complicated conditions. Among these databases, RDB is still mainstream for reasons such as high data management safety and ease of use when applied to application systems.

  As an example of a large-scale database that is searched under relatively complicated and diverse conditions, there is a database (hereinafter referred to as “receipt DB”) of medical fee details (hereinafter referred to as “receipt”). As the receipt DB, for example, in addition to what is managed by each hospital, a national receipt database (hereinafter referred to as NRDB) in which national receipts are aggregated is constructed. NRDB accumulates more than 1 billion receipt data annually. Then, in accordance with a request received from a medical institution or the like, data in a necessary range is extracted from the NRDB and provided.

  When utilizing such a large-scale medical-related database for medical research or national / regional medical / health policy studies, errors in searching and counting can lead to a large number of drugs, vaccines, etc. It may be a waste or shortage factor. For example, when the amount of vaccine stock in a certain region is estimated from the results including the surrounding regions up to the previous year, there is a possibility that a shortage of vaccines may occur if the results are incorrectly searched or tabulated. Even when a search error is noticed and a search (data extraction) is performed again, if the target database is huge, the re-search itself increases the time, processing and monetary costs.

  As a method for performing a correct search in consideration of human error, for example, a search expression written in SQL, which is a standard language for database operations (hereinafter referred to as an SQL sentence) is automatically generated and correct. There is a way to make it possible to write search expressions more reliably. For example, in Patent Document 1, an SQL sentence is automatically generated by concealing an SQL sentence from the user and the user performs an input operation from a simple input screen, and even a user who does not know SQL at all can easily search the RDB. A method is disclosed.

  As another method for performing a correct search in consideration of human error, for example, there is a method of reusing a correct search result obtained in the past or a search expression obtained from the result. FIG. 16 is a block diagram showing a functional configuration of the database utilization system described in Patent Document 2. As shown in FIG. For example, in the method described in Patent Document 2, a user usage table is generated based on data obtained according to a search condition designated by an analysis user, and can be reused by a general user at the judgment of the analysis user. I have to. Thereby, the general user can easily perform an accurate search confirmed by the user in charge of analysis. In the method of Patent Document 2, since a general user can arbitrarily select from a set of generated user usage tables and input a search condition value in a necessary item, the general user can only use SQL. You can search correctly without knowing the data structure.

JP 2004-126680 A Japanese Patent No. 4700462

  In the techniques shown in Patent Document 1 and Patent Document 2, it is not possible to rewrite other than the value of the specific part on the database specified by the search expression, and existing in a similar search in which only part of the structure of the search expression is different. It is not possible to divert the search expression. That is, there is a problem that the scenes in which an existing search expression can be reused is very limited. In addition, the value described in the search condition and its granularity may be too specialized or too fine for the user depending on the application, and there is a problem that selection errors of values and omission of enumeration occur.

  Therefore, an object of the present invention is to provide a database query statement analysis apparatus that can efficiently create an accurate and highly reusable database query statement from an existing database query statement.

  A database query statement analysis apparatus according to the present invention inputs a database query statement, and a user queries the database query at least one of a description in a predetermined form and a conditional expression whose complexity exceeds a predetermined level. To change the database query sentence to a database query sentence from which the burden factor has been removed according to the type of the burden factor, and a burden factor detection unit that detects the burden factor that becomes a burden when the sentence is reused And a change information generation unit for generating the change information.

  According to the database query statement analysis method of the present invention, a database query statement is input, and a user makes a query about at least one of a predetermined form description and a conditional expression whose complexity exceeds a predetermined level. Detected as a burden factor that becomes a burden when reusing a sentence, and generates change information for changing the database query statement to a database query statement from which the burden factor is removed according to the type of the burden factor It is characterized by doing.

  The database query statement analysis program according to the present invention inputs a database query statement to a computer, and the user inputs at least one of a description in a predetermined form and a conditional expression whose complexity exceeds a predetermined level. A burden factor detection process that detects a burden factor that becomes a burden when the database query sentence is reused, and the database query sentence is converted into a database query sentence from which the burden factor is removed according to the type of the burden factor. A change information generation process for generating change information for changing is executed.

  According to the present invention, an accurate and highly reusable database query statement can be efficiently created from an existing database query statement.

It is a block diagram which shows the structure of the database query sentence analysis apparatus in 1st Embodiment by this invention. It is explanatory drawing which shows an example of the database query sentence input into the database query sentence analysis apparatus in 1st Embodiment by this invention. It is a block diagram which shows the structure of the burden factor detection part of the database query sentence analysis apparatus in 1st Embodiment by this invention. It is a block diagram which shows the structure of the change information generation part of the database query sentence analysis apparatus in 1st Embodiment by this invention. It is a flowchart which shows operation | movement of the database query sentence analysis apparatus in 1st Embodiment by this invention. It is explanatory drawing which shows an example of a change information table. It is explanatory drawing which shows an example of change information. It is explanatory drawing which shows an example of a search formula with high reusability, and a table used for it. It is a block diagram which shows the structure of the database query sentence analysis apparatus in 2nd Embodiment by this invention. It is a block diagram which shows the structure of the query sentence change part of the database query sentence analysis apparatus in 2nd Embodiment by this invention. It is a block diagram which shows the example of the structure of the supplementary information input part of the database query sentence analysis apparatus in 2nd Embodiment by this invention, and the displayed user interface. It is explanatory drawing which shows an example of a search type | mold template, a conversion table, and search object description. It is explanatory drawing which shows an example of the database search user interface which can be produced by utilizing the database query sentence analysis apparatus in 2nd Embodiment by this invention. It is explanatory drawing which shows the hardware structural example of the database query sentence analysis apparatus in 1st Embodiment and 2nd Embodiment by this invention. It is a block diagram which shows the principal part of the database query sentence analysis apparatus by this invention. It is a block diagram which shows the structure of the database utilization system described in patent document 2.

  Embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1. FIG.
FIG. 1 is a block diagram illustrating a configuration of a database query statement analysis apparatus according to the first embodiment (embodiment 1). FIG. 2 is an explanatory diagram illustrating an example of a database query sentence input to the database query sentence analysis device. As shown in FIG. 1, the database query statement analysis device of this embodiment includes a burden factor detection unit 11 and a change information generation unit 12.

  The burden factor detection unit 11 inputs a database query sentence (hereinafter referred to as a search expression) as shown in FIG. 2, and when the input search expression is reused, the change according to the reuse purpose is performed. At least one of a difficult part and a part that easily causes an error at the time of change is detected as a burden factor.

  The change information generation unit 12 generates change information for changing the search formula to a search formula with a reduced load factor according to the type of the load factor.

  FIG. 3 is a block diagram illustrating a configuration of the burden factor detection unit 11. As shown in FIG. 3, the burden factor detection unit 11 includes an element description burden analysis unit 111 and a structure description burden analysis unit 112.

  The element description burden analysis unit 111 detects, as a burden factor, when the description of a database table, field, field value, or a set thereof is a burden on the user when reusing a database query statement. . The element description burden analysis unit 111 includes a dedicated code detection unit 1111 that detects, as a dedicated code, a character string that is difficult for a general person to understand, such as a number, a symbol string, and a technical term. ) Are enumerated using an AND clause, an OR clause, or an IN predicate, an element set detection unit 1112 that detects a set of these values as an element set is included. The element description burden analysis unit 111 includes a caution description detection unit 1113 that detects a description of a cautionary part that is likely to cause an error in the use of a database value or the way of writing SQL.

  The structural description burden analysis unit 112 detects a description pattern (frequent pattern) that has a high structural complexity and appears in many search expressions. The structure description burden analysis unit 112 includes a complexity evaluation unit 1121 that evaluates the complexity of a search expression, a conditional expression abstraction unit 1122 that abstracts a conditional expression as necessary, and a conditional expression that stores the abstracted conditional expression A repository 1124 is included. The structure description burden analysis unit 112 includes a frequent pattern extraction unit 1123 that extracts common patterns (frequent patterns) that frequently appear based on the abstracted conditional expressions and the conditional expressions in the conditional expression repository 1124.

  FIG. 4 is a block diagram illustrating a configuration of the change information generation unit 12. The change information generation unit 12 refers to the output result of the burden factor detection unit 11 and the change information table, specifies the change type of the search formula and the necessary change information, and configures the change information configuration unit 121 that configures the necessary change information. A change information output unit 122 that outputs change information for each type of configured burden factor, and a change information table 123 that stores the change information.

  Next, the operation of the database query statement analysis apparatus of this embodiment will be described. FIG. 5 is a flowchart showing the operation of the database query statement analysis apparatus of this embodiment.

  The dedicated code detection unit 1111 detects these character strings as dedicated codes when the input search expression uses character strings that are difficult for the general public to understand, such as numbers, symbol strings, and technical terms. (Step S01). The element set detection unit 1112 detects a set of these values as an element set when values of specific fields in the input search expression are listed using an AND clause, an OR clause, or an IN predicate ( Step S02). The caution description detection unit 1113 detects, as a caution description, a description of a cautionary part that is likely to cause an error in the input search formula using a database value or writing SQL (step S03). The element description burden analysis unit 111 outputs at least one of these dedicated code, element set, and caution description as a burden factor. Further, the field corresponding to the element set, and the description corresponding to the dedicated code and the caution description are predetermined by the system administrator or the like.

  The dedicated code, element set, and cautionary description may be detected from the same part in the search expression. For example, in the example of the search expression shown in FIG. 2, there is a description “prescription.atc in ('N06A3', 'N06A4', 'N06A9')” using an IN predicate in the conditional expression expressed below the WHERE. is there. In this description, for example, 'N06A3' corresponds to a dedicated code, and is also an element set in which three dedicated codes are listed. Furthermore, in the description of the condition “prescription.value> = 30”, the number “30” is a target that is difficult to understand and should be detected as a cautionary description. Note that a DATE type value such as the value of “patients.date” is easy to understand and has low ambiguity, so it is not a cautionary description.

  Note that the order of steps S01 to S03 is not limited to the above order, and may be performed in another order. Further, step S01 to step S03 may be performed in parallel.

  The complexity evaluation unit 1121 performs search expression complexity evaluation (step S04). As a simple method, the complexity evaluation of the search expression by the complexity evaluation unit 1121 can be performed by, for example, the type of predicate used in the conditional expression expressed below or the number of referenced fields and values. Can be realized by determining that the search expression is complicated.

  If it is determined in step S04 that the complexity of the search expression is greater than or equal to a predetermined threshold, the conditional expression abstraction unit 1122 abstracts the conditional expression included in the search expression as necessary (step S04). S05). The conditional expression abstraction unit 1122 adds the abstracted conditional expression to the conditional expression repository 1124.

  The abstraction of the conditional expression by the conditional expression abstraction unit 1122 can be realized, for example, by setting the condition value to be arbitrary (wild card) when the condition includes that a certain field has a specific value. Also, the conditional expression abstraction by the conditional expression abstraction unit 1122 is related to, for example, a table or field with a low appearance frequency, leaving only AND clauses related to a predetermined number of tables and fields having a higher appearance frequency when a plurality of AND phrases are arranged. This can be realized by a method of deleting the AND phrase.

  Next, if a predetermined number or more of conditional expressions similar to the abstracted conditional expression exist in the conditional expression repository 1124, the frequent pattern extraction unit 1123 generates a common pattern (frequent pattern) from the conditional expressions. Extract (step S06).

  A similar conditional expression in frequent pattern extraction by the frequent pattern extraction unit 1123 is, for example, a tree structure in which abstract conditional expressions stored in the conditional expression repository 1124 are AND and OR logical operators as intermediate nodes. (The highest node is WHERE), which is a conditional expression in which the similarity between tree structures is equal to or greater than a predetermined threshold. The similarity between tree structures is, for example, the product of the ratio of common intermediate and terminal nodes and the ratio of attributes (tables and fields) specified in each terminal node (predicate). Appears.

  Through the above processing, the burden factor detection unit 11 outputs, as burden factors, a dedicated code that is difficult for the user to understand, an element set that is likely to cause an excess or deficiency of elements or an entry error when enumerating, and other sensitive parts that are susceptible to error. To do. And the burden factor detection part 11 outputs the frequent pattern showing the part with a high usage frequency of the conditional expression with complexity more than a predetermined | prescribed level as a kind of burden factor.

  The change information generation unit 12 reduces a burden factor from the input search formula according to the type of the burden factor detected by the burden factor detection unit 11 and changes a description content that can be easily reused. Generate information. FIG. 6 is an explanatory diagram illustrating an example of the change information table 123 used by the change information generation unit 12. The change information configuration unit 121 refers to the output result of the burden factor detection unit 11 and the change information table 123, identifies the search formula change type and necessary change information, and configures necessary change information (step S07).

  Next, the change information output unit 122 outputs change information for each type of configured burden factor (step S08). For example, a description “prescription.atc in ('N06A3', 'N06A4', 'N06A9')” including a dedicated code and an element set as a burden factor detected by the burden factor detection unit 11 from the search formula shown in FIG. Referring to the change information table, the change type is a conversion table type, and necessary change information is specified as conversion table information and conversion table embedded type, respectively.

  FIG. 7 is an explanatory diagram illustrating an example of change information generated by the change information generation unit 12. The change information shown in FIG. 7 includes the original search expression R01 and change type R02, and further includes conversion table information R03, conversion table embedded expression R04, alert sentence R05, and conditional expression template R06 as information corresponding to the change type R02. Includes each. Further, in the original search expression R01, a part corresponding to the conversion table information is clearly indicated (underlined part in FIG. 7).

  The change type R02 in the example shown in FIG. 7 includes three types of conversion table type, alert type, and conditional expression template type. As described above, a plurality of change types may be detected from one search expression.

  The conversion table information R03 describes a table name (table.name), a field name (field1.name, etc.), and their values. For some of the values, default values (labels, etc.) are set, and for some of the values, values (atc, N06A3, etc.) included in the burden factor detected from the search formula are set by the change information generation unit 12 The

  Also, the conversion table embedded expression R04 is for replacing a part of the search expression with the description content, and includes an element of the conversion table information R03 in a part of the description content (atc, labels, name, _field1 .value_, where _field1.value_ is a variable that points to the value part of field1.value).

  The alert sentence R05 is generated using a template (not shown) for each type of alert provided in the change information generation unit 12. For example, “prescription.value” and an ambiguous value “30” are detected from “prescription.value> = 30” in the search expression shown in R01 of FIGS. 2 and 7, and these are inserted into the template. As a result, an alert sentence R05 is generated.

  The conditional expression template R06 shown in FIG. 7 includes “OR prescription.atc = 'N06A6'” and “AND prescription.value> = 30” among the conditional expressions (descriptions below “WHERE”) in the search expression shown in FIG. The portion other than (excluding individual field values) is extracted as a frequent pattern included in many other conditional expressions. In the conditional expression template R06 shown in FIG. 7, the conditional expression abstracting unit 1122 of the burden factor detecting unit 11 converts each field value into a wild card (“*”) or a temporary common value (step S05 in FIG. 5). This is a result of the frequent pattern extraction unit 1123 extracting only combinations of descriptions having a high appearance frequency in the WHERE sentence.

  FIG. 8 is an explanatory diagram showing an example of a highly reusable search expression R08 and a conversion table R07 used therefor. The search expression R08 and the conversion table R07 shown in FIG. 8 are obtained by using the change information generated and output by the change information generation unit 12.

  The conversion table R07 output by the change information generation unit 12 is a table in which a new table is defined in the database based on the conversion table information R03, or a record is added to the defined table.

  The conversion table R07 illustrated in FIG. 8 is generated using the conversion table information R03 among the change information illustrated in FIG. In the conversion table R07, the table name is set to “labels”, the first field name is set to “name”, and the second field name is set to “atc”. In the conversion table R07, “N06A3”, “N06A4”, “N06A5”, “N06A9” are set as the values of the second field, respectively, and all the values of the first field are set as “antidepressant”. Has been.

  In the search expression R08 shown in FIG. 8, the conditional expression in the search expression of FIG. 2 is replaced with a conditional expression template R06. In the search expression R08, the change table embedded expression R04 is inserted in the value specification portion of the IN predicate that is a part of the conditional expression template R06. The search expression R08 corresponds to the conversion table R07, the “_field1.value_” portion of the conversion table embedded expression R04 inserted in the value specification portion of the IN predicate in the conditional expression is the common value “name field of the labels table” It has been replaced by an antidepressant.

  Note that by replacing the conditional expression part of the search expression R01 with the conditional expression template R06, the alert statement R05 is unnecessary because there is no condition description to be processed.

  As described above, by creating the conversion table R07 and the search expression R08 from the change information, the user needs to understand a difficult-to-understand dedicated code such as “N06A3”, but what is expressed by listing them multiple There is no need to understand, and instead, the condition can be understood by one easy-to-understand generic term “antidepressant”. Thereby, it can be easily determined whether the existing search formula is a search formula based on a desired condition. In addition, if the generic name of another drug (for example, “diuretic drug”) is defined in the conversion table as well as “antidepressant drug”, only one value “antidepressant drug” is replaced with the generic name of the other drug. Thus, another similar search expression can be easily created.

  In the database query statement analysis apparatus according to the present embodiment, the burden factor detection unit 11 analyzes an existing database query statement, and when a user rewrites a search condition value or configuration and reuses it, a part or an error accompanied by difficulty is detected. Detect a part (burden factor) that is likely to occur. In addition, the change information generation unit 12 generates information necessary for appropriately rewriting the database query statement in accordance with the burden factor type. Therefore, it is possible to easily and efficiently create a database query statement suitable for reuse, which can be searched correctly while avoiding various human errors.

  The element description burden analysis unit 111 included in the burden factor detection unit 11 detects these values when a set of a plurality of values is included in the search condition, and the change information generation unit 12 The conversion table information R03 and the conversion table embedding formula R04 are output so that a set of values can be designated by one generic name (representative value). For this reason, it is easy to specify the value of the search condition, and it is possible to reduce the possibility of description error by specifying a plurality of values.

  Further, the structure description burden analysis unit 112 included in the burden factor detection unit 11 detects a description pattern (frequent pattern) having a high structural complexity and appearing in many search expressions, and the change information generation unit 12 performs the frequent occurrence. The pattern is output as a conditional expression template R06. For this reason, in many situations, it is not necessary for a searcher to write a complicated conditional expression by himself / herself, and an accurate condition can be efficiently specified using a conditional expression template.

Embodiment 2. FIG.
Next, an example of a database query statement analysis apparatus that creates a highly reusable database query statement using change information will be described with reference to the drawings. FIG. 9 is a block diagram illustrating a configuration of the database query statement analysis apparatus according to the second embodiment (second embodiment). The database query statement analysis device of this embodiment includes a burden factor detection unit 11, a change information generation unit 12, a query statement change unit 13, and a supplementary information input unit 14. The burden factor detection unit 11 and the change information generation unit 12 illustrated in FIG. 9 are the same as those in the first embodiment illustrated in FIG.

  Based on the change information as shown in FIG. 7, the query statement changing unit 13 changes the existing search formula to a search formula with high reusability with reduced burden factors, and is further necessary for the post-change search formula. Generate accompanying information. FIG. 10 is an explanatory diagram illustrating the configuration of the query statement changing unit 13 of the database query statement analysis device according to the present embodiment. The query statement change unit 13 includes a condition conversion table generation unit 131 and a conditional expression template application unit 132.

  When the conversion type is included in the change type in the change information, the condition conversion table generation unit 131 generates a conversion table for the value of the conditional expression using the conversion table information R03 in the change information. The method for generating the conversion table is the same as the method for generating the conversion table R07 (labels table) shown in FIG. 8 from the conversion table information R03 shown in FIG. 7 described in the description of the first embodiment. Do it automatically. However, the value specification of the name field of the conversion table R07 (labels table) is performed by the supplementary information input unit 14 described later.

  The conditional expression template application unit 132 applies the conditional expression template and the conversion table embedding expression in the change information to the original search expression, thereby generating an expression that is a base of the search expression with high reusability. For example, a statement in which the conversion table embedded expression R04 in FIG. 7 is inserted in the value specification part of the IN predicate of the WHERE statement shown in the conditional expression template R06 in FIG. 7 and the WHERE statement in the search expression R01 is replaced with the WHERE statement is as follows: Output as a base expression for highly reusable search expressions.

  The supplementary information input unit 14 includes a user interface for the user to enter further necessary information with respect to the conversion table generated by the query statement change unit 13 and a formula that is a base of a highly reusable search formula. . FIG. 11 is an explanatory diagram illustrating an example of a configuration of a supplementary information input unit and a displayed user interface according to the present embodiment. The supplementary information input unit 14 includes a supplementary information input user interface 141 including a search expression operation unit 1411, a conversion table operation unit 1412, and a search target explanation input unit 1413, and a term dictionary 142 that holds supplementary information candidates.

  The search formula operation unit 1411 displays a formula that is a base of the above-described search formula with high reusability, and points out a portion in which a value should be input in advance. In FIG. 11, the value part of the name field included in the value specification part of the IN predicate in the search expression is indicated by a gray box, and the user is prompted to specify some value in the box.

  The conversion table operation unit 1412 provides support for specifying a value in any field of the conversion table described above. In the example shown in FIG. 11, the Japanese notation (such as “mood stabilizer”) corresponding to each value (such as “N06A3”) in the atc field is displayed in advance in the name field of the conversion table using the term dictionary 142. . The user determines, for example, a common value of each displayed name field value as a representative value (such as “antidepressant”) via the conversion table operation unit 1412, and sets all the corresponding name field values. Replace with a representative value. Furthermore, the user also enters the representative value as a search condition in the value of the name field in the search expression requested by the search expression operation section 1411 via the conversion table operation section 1412, so that FIG. The same conversion table (labels table) as shown and a search expression with high reusability can be obtained.

  FIG. 12 is an explanatory diagram illustrating an example of the search formula R08, the conversion table R07, and the search target description R09. As shown in FIG. 12, a search target explanation R09 is given to the conversion table R07 and the search expression R08, and is shown to the user. In the conversion table R07 and the search expression R08, the value of the name field is replaced with a representative value (“antidepressant”) by the operation of the search expression operation unit 1411 and the conversion table operation unit 1412. The user can search for necessary information simply by inputting “year / month / day 1” and “year / month / day 2” of the search target description R09 as a search condition via the search target description input unit 1413. In this way, the user can easily find and use a search expression that is highly reusable.

  FIG. 13 is an explanatory diagram illustrating an example of a database search user interface that can be created by using the database query statement analysis apparatus according to the present embodiment. A user interface for database search as shown in FIG. 13 can be easily created using a highly reusable search expression and conversion table generated by the query statement change unit 13. As shown in FIG. 13, the search condition value can be specified by an easy-to-understand word such as “antidepressant” instead of a set of difficult-to-understand dedicated codes such as “N06A3” and “N06A4”.

  If a similar conversion table is obtained for other than “antidepressants”, “antidepressants” are displayed as alternatives for “antidepressants” in FIG. It is easy to perform a correct search without changing the search expression by the user. In this way, various search expressions accumulated in the past are redefined in a form that can be easily reused, and a database search GUI (graphical user interface) as shown in FIG. 13 based on a search expression including a frequent pattern is used. ) Can provide a means for easily performing a search that is actually frequently used.

  In the database query statement analysis apparatus according to the present embodiment, the query statement change unit 13 uses the information generated by the change information generation unit 12 to change an existing database query statement to a content that can be easily reused, and specifies it as necessary. A conversion table for designating a set of values as representative values is generated. Then, the supplementary information input unit 14 can further easily add supplementary information to the changed database query sentence. This makes it possible to obtain a database query template that is unlikely to cause an error in rewriting or value setting at the time of reuse from an existing correct database query.

  Next, a specific example of the hardware configuration of the database query statement analysis apparatus in the first embodiment and the second embodiment will be described. FIG. 14 is an explanatory diagram illustrating an example of a hardware configuration of the database query statement analysis apparatus according to the first and second embodiments of the present invention. A database query statement analysis apparatus A shown in FIG. 14 includes at least a CPU (Central Processing Unit) A1, a main storage unit A2, an output unit A3, an input unit A4, and an auxiliary storage unit A6. In addition, the database query statement analysis apparatus A may include a communication unit A5.

  The main storage unit A2 is a main memory such as a RAM (Random Access Memory), for example, and is used as a data work area or a temporary data save area. The output unit A3 is a display device such as a liquid crystal display device or a printing device such as a printer, and outputs data. The input unit A4 is an input device such as a keyboard or a mouse and inputs data. When data is input by reading a file, the input unit A4 may be an external recording medium reader. The auxiliary storage unit A6 is, for example, a ROM (Read Only Memory) or a hard disk device. Further, as shown in FIG. 14, in the database query statement analysis apparatus A, the above constituent elements A1 to A6 are connected to each other via a system bus A7.

  In addition, the auxiliary storage unit A6 stores programs for realizing the burden factor detection unit 11, the change information generation unit 12, the query statement change unit 13, and the supplementary information input unit 14 illustrated in FIG. 1 or FIG. Further, the auxiliary storage unit A6 includes a conditional expression repository that is read and written by the burden factor detection unit 11, a change information table 123 that is referred to by the change information generation unit 12, a change information that is output by the change information generation unit 12, and a supplementary information input unit 14. Stores data representing the search expression R08 and conversion table R07 output by

  The input unit A4 or the communication unit A5 acquires a database query statement. Further, the supplementary information input unit 14 according to the second embodiment makes an inquiry for prompting the user of the database query statement analysis apparatus A to input supplemental information through the output unit A3.

  It should be noted that the database query statement analysis apparatus A has a circuit composed of hardware components such as LSI (Large Scale Integration) in which a program for realizing the functions shown in FIG. 1 or FIG. 9 is incorporated. Thus, it may be realized by hardware. Further, as shown in FIG. 14, the program that provides the function as shown in FIG. 1 or FIG. 9 may be executed by software by causing the CPU A1 of the computer to execute the program. In this case, the CPU A1 loads the program stored in the auxiliary storage unit A6 into the main storage unit A2, executes it, and controls the operation of the database query statement analysis apparatus A, thereby enabling the above functions to be performed by software. Can be realized.

  The communication unit A5 is connected to a peripheral device and has a function of transmitting / receiving data. As one of the peripheral devices, the external storage device B shown in FIG. 14 is connected to the database query statement analysis device A via the network (information communication network) by the communication unit A5 and stored in the external storage device B. May be acquired by the burden factor detection unit 11. Furthermore, the change information data output from the change information generation unit 12 described above, the search expression template output from the supplemental information input unit 14, and data representing a conversion table may be stored in the external storage device B.

  FIG. 15 is a block diagram showing the main part of the database query statement analysis apparatus of the present invention. As shown in FIG. 15, the database query statement analysis apparatus inputs a database query statement, and at least one of a description in a predetermined form and a conditional expression whose complexity is equal to or higher than a predetermined level is input to the user. A burden factor detecting unit 11 that detects a burden factor that becomes a burden when the database query statement is reused, and a database query in which the burden factor is removed according to the type of the burden factor And a change information generation unit 12 that generates change information for changing to a sentence.

  Moreover, in said embodiment, the database query sentence analysis apparatus as shown to the following (1)-(5) is also disclosed.

(1) A database query statement is input, and a description in a predetermined form (for example, a description detected from a dedicated code detection unit 1111, an element set detection unit 1112, a caution description detection unit 1113) and a complexity are predetermined. A burden that is a burden when the user reuses the database query statement with at least one of the conditional expressions (for example, the conditional expression evaluated to be complicated by the complexity evaluation unit 1121) that is equal to or higher than A burden factor detection unit (for example, a burden factor detection unit 11) to detect as a factor, and a change for changing the database query statement to a database query statement from which the burden factor has been removed according to the type of the burden factor A database query statement analysis apparatus including a change information generation unit (for example, a change information generation unit 12) that generates information.

(2) The database query statement analysis apparatus includes an element description burden analysis unit (for example, an element description burden analysis unit 111) in which a burden factor detection unit detects a part related to designation of information recorded in the database among burden factors. The structure description load analysis unit (for example, the structure description load analysis unit 112) that detects a part related to the structure of the database query sentence among the burden factors may be included.

(3) In the database query statement analysis apparatus, the element description burden analysis unit has a database query statement in which a set of a plurality of values of the same field recorded in the database (for example, a set of values detected by the element set detection unit 1112). Is included as a search condition, the set of the plurality of values is detected as a burden factor, and the change information generation unit can specify the set of the plurality of detected values as one representative value. The information (for example, conversion table information R03) necessary for defining the conversion table indicating the correspondence between the set of the plurality of values and the representative value may be output as the change information. According to such a database query statement analysis apparatus, it is easy to specify the value of the search condition, and it is possible to reduce the possibility of description error by specifying a plurality of values.

(4) In the database query statement analysis apparatus, the structure description burden analysis unit (for example, the structure description burden analysis unit 112) includes a type of predicate, the number of fields, and the number of values from a plurality of database query statements. A frequent pattern of a conditional expression whose complexity determined based on at least one of them is equal to or higher than a predetermined level is detected as a burden factor, and the change information generation unit uses the frequent pattern to determine a template for the conditional expression ( For example, the conditional expression template R06) may be generated as change information. According to such a database query sentence analysis apparatus, it is possible to efficiently specify an accurate condition using a conditional expression template without requiring a searcher to write a complicated conditional expression in many situations.

(5) The database query statement analyzer includes a query statement change unit (for example, the query statement change unit 13) that changes the database query statement using the change information generated by the change information generation unit, and the changed database query statement And a supplementary information input unit (for example, supplementary information input unit 14) for inputting supplementary information including at least a representative value used in the conversion table. According to such a database query statement analysis apparatus, a database query statement template that is unlikely to cause an error in rewriting or value setting during reuse can be obtained from an existing correct database query statement.

  The present invention can be applied to a search system in which a database query statement is reused.

DESCRIPTION OF SYMBOLS 11 Burden factor detection part 12 Change information generation part 13 Query sentence change part 14 Supplementary information input part 111 Element description burden analysis part 112 Structure description burden analysis part 131 Condition conversion table generation part 132 Conditional expression template application part 1411 Search expression operation part 1412 Conversion table operation unit 1413 Search target explanation input unit A Database query sentence analysis device A1 CPU
A2 Main storage unit A3 Output unit A4 Input unit A5 Communication unit A6 Auxiliary storage unit A7 System bus B External storage device

Claims (8)

When a database query is entered, at least one of a description in a predetermined form and a conditional expression whose complexity exceeds a predetermined level becomes a burden when the user reuses the database query. A burden factor detector that detects the burden factor;
A database query comprising: a change information generation unit that generates change information for changing the database query statement to a database query statement from which the burden factor has been removed according to the type of the burden factor Sentence analysis device. The burden factor detector
Among the burden factors, an element description burden analysis unit that detects a part related to designation of information recorded in the database;
The database query statement analysis apparatus according to claim 1, further comprising: a structure description burden analysis unit that detects a portion related to a structure of the database query statement among the burden factors. The element description burden analysis unit
When a set of multiple values of the same field recorded in the database is included as a search condition in the database query statement, the set of multiple values is detected as a burden factor,
The change information generation unit defines a conversion table that represents a correspondence relationship between the set of the plurality of values and the representative value so that the detected set of the plurality of values can be designated by one representative value. The database query statement analysis device according to claim 2, wherein information necessary for the output is output as change information. The structure description burden analysis unit has a complexity determined based on at least one of the type of predicate, the number of fields, and the number of values from a plurality of database query statements, and a predetermined level or more. Detect frequent patterns of conditional expressions as burden factors,
The database query statement analysis device according to claim 2, wherein the change information generation unit generates a fixed template of a conditional expression as change information using the frequent pattern. The structural description burden analysis unit detects a frequent pattern of conditional expressions having a complexity of a predetermined level or more as a burden factor from a plurality of database query statements.
The database query statement analysis device according to claim 3, wherein the change information generation unit generates a fixed template of a conditional expression as change information using the frequent pattern. A query statement change unit that changes the database query statement using the change information generated by the change information generation unit;
The database query statement analysis apparatus according to claim 3, further comprising a supplementary information input unit that inputs supplementary information including at least a representative value used in the database query statement and the conversion table that have been changed. When a database query is entered, at least one of a description in a predetermined form and a conditional expression whose complexity exceeds a predetermined level becomes a burden when the user reuses the database query. Detected as a burden factor,
A database query statement analysis method, comprising: generating change information for changing the database query statement to a database query statement from which the burden factor has been removed according to a type of the burden factor. On the computer,
When a database query is entered, at least one of a description in a predetermined form and a conditional expression whose complexity exceeds a predetermined level becomes a burden when the user reuses the database query. Burden factor detection processing to detect as a burden factor;
A database query statement analysis program for executing change information generation processing for generating change information for changing the database query statement to a database query statement from which the burden factor has been removed according to the type of the burden factor .