近年來,專利/技術之重要性乃逐漸提升,專利授權知識之相關應用乃漸為廣泛,如專利/技術授權之價格鑑定機制乃需仰賴專利授權資訊,以作為價格鑑定機制判斷之指標;是故,專利授權知識對於專利需求者(如專利鑑價者、專利發明者等)之重要性乃逐漸增加。隨著專利授權知識之應用日趨重要,專利授權知識之擷取、儲存、管理與再利用乃為業界所關注之重要課題。然而,就專利授權領域所存在之知識管理相關議題而言,過去研究甚少融入自然語言處理與知識表達等技術進行專利授權文件之剖析,而主要係以人工方式透過各搜尋網站蒐集專利授權相關資料,再以人工閱讀方式擷取所需之專利授權知識,進而歸納、彙整以形成結構化文件,以便進行後續之專利授權資料分享與交換。此外,隨著時間之累積,專利授權領域之網路資源已逐漸增加,並超過人類所能整理與解讀之負荷。有鑑於此,本研究乃針對專利授權領域提出專利授權文件之知識結構解析與表達模式,自該領域文件中自動擷取具有價值且有意義之專利授權知識並予以結構化表達。其中,「專利授權知識結構解析模組」乃以詞彙發生頻率及個別詞彙/詞句資訊(如詞彙/詞句長度、詞彙權重值等)為依歸而發展,並考量各專利授權知識屬性之特性,建構適用之法則,以使解析結果更具正確性。而「專利授權知識表達模組」則以語意網之知識表達概念為基礎,配合RDF語法及知識本體結構技術,使專利授權知識文件之表達具有語意層次之結構,以有效達到專利授權知識之自動化表達、分享與交換之目標。 此外,為便於專利需求者(如專利鑑價者、專利發明者)使用專利授權資訊,並將專利授權資訊應用於專利價格鑑定之指標等用途,本研究乃於網際網路環境下建構一套專利授權知識結構解析與表達系統,提供專利需求者一個自動化處理專利授權文件之環境,進而提升專利授權知識之管理效能與再利用率。
In recent years, importance of patents/advanced technologies has been realized and thus applications of patent licensing knowledge have been stressed. As a result, all typical knowledge management activities (e.g., knowledge extraction, storage, and reuse) for have become the key issues for management of patent licensing knowledge. Traditionally, the contents of patent licensing documents have to be recognized by domain experts. Concerning complexity of different types of patent licensing documents, this paper develops an approach for structure extraction and knowledge representation of the patent licensing documents. The algorithms for knowledge extraction, fragmentation, natural language processing (NLP), knowledge representation and resource description framework (RDF) are applied for automatic and structured representation of patent licensing documents. In addition, a Web-based knowledge management system for patent licensing documents is also developed in order to facilitate exchange of patent licensing knowledge. The aim of this research is to establish an applicable environment for sharing of patent licensing knowledge and enhance the efficiency for patent licensing knowledge management.