Information extraction from semi and unstructured data sources: a systematic literature review

• Main Authors: Zaman, Gohar, Mahdin, Hairulnizam, Hussain, Khalid, Rahman, Atta-ur-
• Format: Article
• Language: English
• Published: ICIC-EL Office 2020
• Subjects: T Technology (General)
• Online Access: http://eprints.uthm.edu.my/6551/1/AJ%202020%20%28348%29.pdf; http://eprints.uthm.edu.my/6551/; https://dx.doi.org/ 10.24507/icicel.14.06.593
• Institution: Universiti Tun Hussein Onn Malaysia

Millions of structured, semi structured and unstructured documents have been produced around the globe on a daily basis. Sources of such documents are individuals as well as several research societies like IEEE, Elsevier, Springer and Wiley that we use to publish the scientific documents enormously. These documents are a huge resource of scientific knowledge for research communities and interested users around the world. However, due to their massive volume and varying document formats, search engines are facing problems in indexing such documents, thus making retrieval of information inefficient, tedious and time consuming. Information extraction from such documents is among the hottest areas of research in data/text mining. As the number of such documents is increasing tremendously, more sophisticated information extraction techniques are necessary. This research focuses on reviewing and summarizing existing state-of-theart techniques in information extraction to highlight their limitations. Consequently, the research gap is formulated for the researchers in information extraction domain.