NLP and cognitive research
The nerve centre of the technology is Natural Language Processing (NLP), which makes it possible to read, decipher and make sense of human language in an evaluable way.
Natural language processing consists of a succession of stages: 1) Tokenization; 2) Morphological and lexical analysis; 3) Syntactic analysis and parse tree generation; 4) Semantic and discourse analysis.
Lisia uses cognitive search systems , a new generation of search tools that output more relevant results for the user.
Through this process we improve the relevance of the results and automatically optimise the relevance of the results over time. Algorithms continuously refine and improve search results, providing the best relevance to users.
How it works
The Lisia algorithm uses a semantic/vector system to, in a fraction of a second, sort millions of legal data contained in full sentences and maxims in our Search Index to find the most useful and relevant results for what you are looking for.(Information retrieval).
In full sentences, the system is able to highlight the relevant part of the sentence by automatic scrolling (text extraction & automatic highlighter).
Through these tools, the search focuses only on the relevant and pertinent part of the documents, speeding up and simplifying reading.
Users want quick answers to their queries. We have made a lot of progress to provide the most relevant answers, in the fastest way and in the most useful ways.
We rely on extensive testing and have an automated evaluation process to analyse metrics and decide whether to implement a proposed change or feedback. The data from these evaluations and experiments is carefully reviewed by experienced engineers and research analysts, as well as lawyers who determine the decision.
Our team of engineers and lawyers have many ideas for the future to make Lisia's results even more useful and products even more innovative.