ProDig DataSearch - created for enterprise

Simple search "find everything, where is this word" in company often gives you thousends documents, which is almost useless.

Classical search (Google) are optimalized for web pages, where the main goal is to find the most relevant web page from large amount of web pages with different quality. Its ranking is based mostly on hyperlinks. In company environment are all documents relevant, their content is very similar and there are no hyperlinks at all. It is not possible to discover automatically logical bonds between documents. That's the reason why most advantages of these search engines take no place in company environment.

DataSearch is optimalised exactly for company environment. It is able to constrain search just for expected types, logical position/location (context), time interval, connect outputs with database (e.g. accounting software) and line them up according many criteria, including date or indications directly from searched documents (e.g. author's name, sum in balance sheet).

Distinguish logical emplacement in data

Most document's formats was created for human processing, that's why they have a lot of possibilities for shape adjustment (writing, colors, tables, etc.), but just few for context description (name, address, ID and alike). People doesn't have problems to recognize, that three rows in the upper right side in document is address, or that number in the brackets behind company's name is its ID, but for a computer it a big problem. In addition formating rules are very heterogeneous, making a software able to do it universally is almost impossible.

ProDig DataSearch thanks to wide configuration options and cooperation with ProDig DataImport is able to set up options exactly for your data, which you need to process. It can "learn" to distinguish all context components, that you are interested in the search process and make it much easier and faster to find necessary information.

Assignments like "find word 'oil' just like company name", "find '123' just like part of phone number" or "find 'beautiful' just like part of address and only on company purchase orders from Bratislava" are with DataSearch no problem.

Documents as well as database

Database was in history used only for "table" data saving, in which has fulltext search just limited utilization. But today it is used more and more often to save text data or even whole documents. In addition even in table data are extending informations eligible to use in the search process as well.

ProDig DataSearch contains complete support for database systems and allows fulltext search as well as merge "classical" access into database (SQL) with fulltext.

Thanks to this feature query "Zentiva" automatically finds also "Slovakofarma" (old company name) and doesn't have problems with queries like "FIXME articles about all companies with business relation and liabilities over million".

Arrangement and filtering

Arranging search outputs seems like trivial problem but it is technicaly very difficult. Google is able to arrange just according relevance (only in case of gruops also according date of addition). For company environment it is exactly immportant to arrange data also according different criterias.

DataSearch for arranging permits to use context recognition a line up according almost any context criteria directly in document - e.g. maturity date, company name, value of owners equity, etc.

All arranging criterias can be at the same time used for filtering, for example "just companies with 100 to 500 employees" or "just documents refering to issues in may 2006". This filtering is of course able to combine itself with any search criterias.