The Four Objectives of Data Processing

• Data conversion
  Data gathered and stored in one format may need modified to ensure compatibility with other software, systems and data. Most data formats can be mapped provided the original and destination formats are well defined. In exceptional circumstances the requirement for a clear description of the original format can be waived. We have on occasion reconstructed and converted raw disk dumps of library databases with no initial knowledge of the data structures involved.
  
  
• Data modification
  Sometimes datasets need updating, cleaning or merging with other datasets. This can be automatically performed in-house by bespoke software written by us, or the entire process can be given to Protechnix to implement.
  
  
• Data enhancement
  Data of mediocre quality can be improved either by computational transformations of the dataset itself or via intersection with data of higher quality.
  
  
• Data analysis
  ProTechnix can statistically investigate vast quantities of data using high speed analysis tools written in C and C++. High speed statistical tools are particularly appropriate when studying data warehouses for significant trends and features. We also write custom tools for rare or new statistical techniques. Examples include :
  
  
  • Univariate adaptive spline autoregression
  • Complex domain neural networks
  • Wavelet transforms
  • MARS
  • Gabor transforms
  • Weiner filtering
  • Dimension estimation using delay coordinate embedding

   

  If you have a requirement for one of these data services please contact Protechnix at

In addition to general data processing we also provide specialist bibliographic data processing services for public and private libraries. Click here for further information.