Patent attributes
Technical solutions are described for improving the performance of natural language processing systems and other such human-computer interaction systems by facilitating analyzing unstructured computer text by converting such unstructured computer text to domain-specific groups using network graphs. The technical solutions use a graph to connect similar terms with attributes and structural information to facilitate the grouping of different terms that may be used to describe the same entity. Technical solutions facilitate analyzing different input data to generate a graph that can be further used to find data similarity in the input data. The generated graph captures attributes associated with each term and assigns groupings for all the terms at the same time, improving the performance of the natural language processing (NLP) system that is analyzing the input data.