DEVELOPMENT OF A FRAMEWORK FOR PROCESSING UNSTRUCTURED TEXT DATASET THROUGH NLP IN COST ESTIMATION AEC SECTOR

Cost estimation is one of the most critical steps in the building construction process. Currently, it requires humans to manually extract information from documents written in natural language, often resulting in human error. This paper aims to investigate an automated technique for extracting data from documents with the support of NLP techniques, in order to automatize the task of structuring information. A framework for automatically classifying information from unstructured text was developed leveraging NER techniques. This research supports the cost estimation activity minimizing the loss of resources coming from human error when interpreting NL documents.