Research on Structured Extraction Method for Function Points Based on Event Extraction

Software size is a significant input for software cost estimation, and the implementation of estimation dramatically affects results efficiency estimation. Traditionally, implemented by strictly trained experts more labor-intensive large projects, which relatively expensive inefficient. Function Point Analysis widely used method supported several international standards. We propose structured automated function point extraction based on event in natural language processing to address problem complex inefficient manual recognition recognition. This approach has been validated 10 industrial cases. Experimental show that our can identify than 70% points, significantly improves implementation. paper could be guide application artificial intelligence techniques