Generation of requests to a data processing system

Abstract

The present invention relates to a method for an at least semiautomatic generation of a request to a data processing system with a data pool, wherein a data model, which is usable for the selection of data from the data pool using a query language, forms the basis of the data pool, wherein a user's request made in natural language, which takes place by a user input of terms (user terms) captured as a series of characters (character string), wherein the terms semantically describe the user intended results of the answer of the data processing system on the request that has to be carried out by the data processing system, is analyzed and/or interpreted, and converted into a request in the query language.

What is claimed is: 
     
         1 . The method for an at least semiautomatic generation of a request to a data processing system with a data pool,
 wherein a data model, which is usable for the selection of data from the data pool using a query language, forms the basis of the data pool,   wherein a user's request made in natural language,
 which takes place by a user input of terms captured as a series of characters (character string), 
   wherein the terms semantically describe the user intended results of the answer of the data processing system on the request that has to be carried out by the data processing system, is analyzed and/or interpreted, and converted into a request ( 9 ) in the query language, characterized in that the request in natural language captured as a character string is resolved into lexemes, those lexemes are sequentially processed equivalent to their sequence,   wherein a meaning of the particular lexeme is determined on the basis of the data model, and a request in the query language is created out of the determined meaning.   
     
     
         2 . The method according to  claim 1 , wherein for each particular lexeme to be processed a check is performed if a subsequent lexeme in the sequence is existing, wherein it is checked in this case, if the lexeme currently to be processed is the description of a data source, the description of a data class, or the description of a data attribute of the data model. 
     
     
         3 . The method according to  claim 2 , wherein in case the description of a data source or the description of a data class is present, the description is transformed into a parameter of the query language for the selection of data from the data pool, wherein the parameter narrows down the selection of data on data with the description of the data source or the description of the data class. 
     
     
         4 . The method according to  claim 2 , wherein the data source is checked only for the first lexeme in the sequence. 
     
     
         5 . The method according to  claim 1 , wherein it is checked if the lexeme currently to be used is a filler word, wherein the lexeme currently to be processed is ignored in this case and the subsequent lexeme in the sequence is used. 
     
     
         6 . The method according to  claim 2 , wherein in case the description of a data class is present, it is checked if the request in the query language contains already a parameter which narrows down the selection of data to data with the description of the data class, wherein in this case the currently to be processed lexeme is transformed to a parameter of the query language for selection of data from the data pool, wherein the parameter describes a relation between data objects, which are described with the parameter already contained in the request using the query language, and data objects, which are described with parameters beginning with the description of the currently to be processed lexeme. 
     
     
         7 . The method according to  claim 2 , wherein in case the description of a data attribute is present, the description and the subsequent lexeme in the sequence are transformed into a parameter of the query language as a name-value-pair for the selection of data from the data pool, wherein the parameter narrows down the selection of data to data with the description of the data attribute. 
     
     
         8 . The method according to  claim 6 , wherein in case the description of a data class is present, it is checked if the description is also the description of a data attribute, wherein in this case the currently to be processed lexeme and the subsequent lexeme in the sequence are transformed into a parameter of the query language as a name-value-pair for the selection of data from the data pool, wherein the parameter alternatively narrows down the selection of data to data with the description of the data attribute, meaning that in the request the OR-conjunction is combined with the request of data of a data class. 
     
     
         9 . The method according to  claim 2 , wherein in case no data source, no data class, no data attribute or no remaining lexeme in the sequence in the character string is described, the currently to be processed lexeme is transformed to a parameter of the query language for the selection of data from the data pool, wherein the parameter narrows down the selection of data to data with the currently to be processed lexeme as a free-text search.