本研究的目的在於發展一個對於非結構或半結構化文件的語意感知機制,其用途為萃取文件中有用的隱含知識,透過企業年報為範例,本研究以此機制分析企業年報中所蘊含的智慧資本內容。由於ㄧ般投資人對於投資標的往往不知如何判定是否具有競爭優勢,只能聽從專業投資人的建議或是以過去企業經營績效為依據來判斷等,如此方式往往無法取得投資先機。再者由於企業的競爭力已漸漸由傳統的機器設備、資金、原料及勞工改為無形的知識資產,例如:經營團隊的領導力、員工向心力、顧客及供應商的關係、創新文化、品牌、及研發能力等。如此更是增加判斷企業競爭力的困難度。ㄧ般人對於上述所謂的無形資產往往只聞其名而不知其義。因此本研究透過整理智慧資本文獻方式建立其知識本體,再透過文字探勘的文件分群技術萃取其特徵詞彙。透過專家的判斷結合這些特徵詞彙成為專家概念,最後以知識本體的語意規則的方式連結此兩者,使一般投資人可以由文件中了解企業所蘊含的智慧資本做為投資參考。
. This study aims to help people invest in some of the companies have a competitive advantage. Because of the competitiveness of enterprises has changed from tangible assets to intangible assets, for example, the value of intellectual capital: goodwill, innovation, brands, etc. People often don't know how to judge the company's competitiveness. While prior studies focus on insights for the relationships between intellectual capital indicators and bussiness performance. Through the text-mining and ontology building technology , this study builds a semantic aware mechanism to extract the intellectual capital information from enterprise's annual reports.