Automatic summarization of documents

• Main Author: Min, Lingduo
• Other Authors: Mao Kezhi
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/64556
• Institution: Nanyang Technological University

Automatic Knowledge Extraction system from unstructured Open Source data (AKEOS) is an artificial intelligent system utilizing machine learning techniques to help users summarize the knowledge and generate the relationships among the important entities to their queries. AKEOS is realized in forms of web application on Python Web framework Django. The web application executes user’s query by analyzing the data crawled from Google searched results, with the help of the implemented text-miming algorithm, hence summarizes the knowledge and constructs the relevant relationships between entities automatically. In the end, the system renders the entity relationships to user intuitively in terms of a diagram with captions. The AKEOS system is composed of six parts: URL Crawler, Sentence Extraction, Feature Words Generation, Relevant Sentence Selection, Entity Extraction and Knowledge Graph Generation. The individual parts works in tandem and eventually generate the graph based on the computed data.