International Journal of Engineering and Technology

Data mining is the exploration and evaluation of large quantity of data to discover substantial, novel, useful and effectively understandable data. Hence determining the knowledge of a document becomes a necessary task in data mining. There are three approaches of metadata in general. They are stylistic, machine learning and knowledge bases. Sometimes the problem occurs when mining a document that contains polysemic words which leads to irrelevant extraction and increased processing time. Polysemy refers to coexistence of many possible meaning for a word or phrase. In order to extract exact information, polysemy like issue should be solved. This work uses knowledge based metadata to extract information using Domain-based Information Extraction technique (DIE). Hence this work targets in solving polysemy which can increases the accuracy of information extraction and reduce processing time. By applying this method to a enormous amount of Engineering domains contains fields like computer science, biomedical, nanotechnology, physics, this work shows that the information extraction is efficient for day-to-day applications with reduced processing time.