Abstract: Automatically classifying product description information includes selecting a first word from the information and determining whether the word is defined as a prefix within one or more keywords, each keyword being associated with one of a plurality of classes, each class being associated with one or more keywords. If the word is defined as a prefix, then for each keyword for which the word is defined as a prefix, determine whether all suffixes of the prefix within the keyword are found among all remaining words in the information in sequence. For each keyword for which this is true, generate a new result for each class associated with the keyword.

Abstract: A PDF file is received from an external application and key information from the PDF file, concerning a PDF document contained in the PDF file, is extracted. The extracted key information is compared with analogous reference information stored for previously received PDF documents to determine a status for the received PDF document, which is provided to the external application from which the PDF file was received. The received PDF document is a duplicate if all the extracted key information matches analogous reference information, is an updated document if certain but not all the extracted key information matches analogous reference information, or is a new document if at least certain extracted key information does not match analogous reference information. If the PDF document is a duplicate document, processing of the PDF file is returned to the application without storing any extracted key information as reference information.

Abstract: A computer-implemented method for mapping multiple source content schemas into a single target content schema. For each source schema to be mapped, a source class tree representing the source schema classes and a target class tree representing the target schema classes are provided for display to a user, input specifying one or more source properties within one or more source classes is received, one or more target properties within one or more target classes are determined for mapping to the source properties, and a schema map file including a mapping of the target properties to the source properties is generated. The source databases are determined. The schema map files are applied to generate information used for extracting the appropriate content from the source databases and for loading the extracted content into the target database.

Abstract: A PDF file is received from an external application and key information from the PDF file, concerning a PDF document contained in the PDF file, is extracted. The extracted key information is compared with analogous reference information stored for previously received PDF documents to determine a status for the received PDF document, which is provided to the external application from which the PDF file was received. The received PDF document is a duplicate if all the extracted key information matches analogous reference information, is an updated document if certain but not all the extracted key information matches analogous reference information, or is a new document if at least certain extracted key information does not match analogous reference information. If the PDF document is a duplicate document, processing of the PDF file is returned to the application without storing any extracted key information as reference information.