Natural Language Processing with Rosoka

Data today comes in many formats. As a result, analysts and investigators not only have to deal with a spreadsheet to analyse, but also with reports, e-mails, social media posts and other forms of unstructured data. Manually sifting through all this information is time consuming and carries with it the risk of overlooking important information. Natural Language Processing (NLP for short) provides the solution for analysing these unstructured data formats.

What is NLP?
Natural Language Processing (NLP) is a form of artificial intelligence (AI). It is concerned with decoding and analysing human language and communication between humans and computers. In NLP, human language is broken into pieces and given a meaning that a computer understands. This allows computers to analyse language as we do ourselves using context, grammar and meaning. Some applications of NLP are: the spam filter in your mailbox, chatbots, GPS devices or, for example, digital assistants with voice recognition like Siri.

Natural Language Processing in Rosoka
Rosoka offers a scalable, multilingual NLP solution suitable for investigative authorities, commercial organisations and financial institutions. The software enables these organisations to gain insights from large amounts of unstructured investigation data or written text. Rosoka automatically extracts and displays the most important entities relevant to an investigation and points out the relationships between them. This immediately provides insight into the who, what, where and when.

Using the ready-to-use extraction solutions within Rosoka, it is then possible to immediately start analysing the results. Moreover, the solution can be deployed in any language, on any platform and on any device or in any cloud.

Different types of techniques
Natural Language Processing is not one type of technique. There are several types of techniques that can be used in our daily activities. Here are some examples:

• Optical Character Recognition (OCR): a technique that allows computers to read letters, numbers and punctuation marks from an image.
• Named Entity Recognition (NER): this algorithm extracts important entities present in the text, such as people's names, money values, and timestamps.
• Machine Translation: this tool helps analyse and generate language. For example, think Google Translate or the automatic translation of posts on social media.
• Sentiment analysis: this tool interprets and classifies emotions indicated in a text.

Want more information about Rosoka? Please contact us.