Discuss the reliability factor of the NLP system? the answer should be with reference

question reference: when we used machine learning or NLP in analyzing financial data what is the reliability of this data.

General Electric's Research and Development Center has developed a natural language system called SCISOR (System for Conceptual Information Summarization, Organization, and Retrieval) that performs text analysis and question-answering in a limited, predefined subject area (called a constrained domain). One application of this system deals with analyzing financial news. For example, SCISOR automatically selects and analyzes stories about corporate mergers and acquisitions from the online financial service of Dow Jones. Ii is able to process the news in less than 10 seconds per story. First, it determines whether the story is about a corporate merger or acquisition. Then, it selects information such as the target, suitor, and price per share. The system allows the user to browse and ask questions such as, "What price was offered for Polaroid?" or "How much was Bruck Plastics sold for?"

The system's effectiveness was demonstrated in testing when it proved to be 100 percent accurate in identifying all 31 mergers and acquisitions stories that were included in a universe of 731 financial news releases from the newswire service.