Text analysis

Companies tend to possess large quantities of highly valuable information scattered around in unstructured documents. Making use of this information can prove to be a very costly task and many companies often balk at the task.

Serikat's technology makes it possible to extract and structure the information automatically for subsequent use.

Capture and labelling, locating the organisations that meet the rules you require.

Cataloguing, clearly identifying key data in the document.

Data anonymization, replacing sensitive data contained in the text with others, so that the text and the knowledge contained can be transferred to the entire organisation.

Association between certain types of documents (identified by locating certain information) using automatic links.

Classification, identifying the document on the basis of its morphological characteristics.

Classification, by means of analysis of the contents, selecting the taxonomies it refers to.

Automated edition of publications, for automating the process of editing and layout of publications by adjusting them to the style book.

Successful case studies

General Council for the Judiciary-Judicial Records Centre: Processing of Court Decisions

The General Council for the Judiciary's objective was to provide the Spanish legal community with a complete set of all relevant decisions (verdicts and rulings) in structured format. It delegated this task to its Judicial Records Centre.
With its semantic technology, SERIKAT contributed the necessary standardisation, structuring, data anonymization and classification, structuring the contents so that they can be processed by knowledge management and big data systems.
We have processed over 5 million decisions.
  • All information that will help with two major challenges – acceleration and standardisation of judicial decisions – made available to judges, magistrates and the legal world in general.
  • Dramatic reduction in document cataloguing and classification costs.
  • Enrichment of the information by extracting data from the texts that goes beyond their contents.
  • Detection of duplicates.
  • Homogenisation of information, giving similar data identical values.
  • Improvement in legal practice, through location and linking of jurisprudence and legislation.
  • Hiding sensitive information, by means of depersonalisation of documents, for compliance with the Data Protection Act and Heredia rules.
You can find us at