Research On Intelligent Information Extraction Approaches For Business Value Evaluation

Due to an exponential increase in information resources on the Internet over the last several years,the birth of the search engine has brought the most convenience and efficiency for users.However,searching data by keyword retrieval techniques in information retrieval is not contented with some users' specific needs due to a large number of network resources on the Internet.Information extraction is an approach which extracts the important specific event or produces specific relations among information from documents.Information extraction can not only filter unnecessary information in any documents but also produce specific important messages and summaries that users are interested in.It becomes an important tool for business value evaluation.Business value evaluation is collecting,analysis,and applying to financial information to appraise the business value.The evaluated results are used in carrying out business activities such as stock listings and merge and acquisition decisions.There are specific information and events about business valuation stored in the Intelligent financial statements,notes to financial statements,and financial news of China companies at present and data is presented by HTML and PDF files.In order to get business value evaluation related financial data quickly and accurately instead of manually collecting these financial data,to develop a suit of intelligent data extraction approaches are highly important.In this paper,we split this research into 4 chapters,the first chapter initially introduces theories and methods for intelligent information extraction,including Chinese words segmentation,POS Tagging,keywords extraction and Table data extraction.Chapter 2 develops exaction methods based on different sources using modern technology and an information extraction system of Chinese financial data for business valuation from the domestic financial statements,notes to financial statements,and financial news as the data sources and give detailed explanation for each method.Chapter 3 designs and develops a prototype system for methods of intelligent financial data extraction,including user interface module,Chinese word analysis and financial data extraction module and applies methods of financial data extraction and knowledge base model into the system,and then introduces three experiments according to different data sources,i.e financial statements,notes to financial statements,and financial news and verify if the proposed approaches work and do analysis and summary on experiment results.Chapter 4 applies the system and intelligent,financial data extraction into business activities,such as,enterprise merger and reorganization,financial risk evaluation,stock investment and etc,in which human resources and cost are saved a lot.This paper highlights:it applies methods of natural language processing and gets the correct financial data and combine their corresponding Business value evaluation Model to achieve an automatic extraction in the financial data from these different heterogeneous data sources.Users can collect correct financial data for business valuation within the shortest time to improve the efficiency in text processing quality.Additionally,optimizing and improvement of POS tagging technology,which greatly improves the exactness.