MULTICLASS RESUME CATEGORIZATION USING DATA MINING

Breakthrough of expeditious digitalization of data accretion caused a stupendous magnitude of data. Nearly 80 percent of data comprises semi-structured and structured data in this alarming era of digitalization. To innovate relevant ways, trends, functions and patterns to analyze the functioning of these are quite difficult and a big deal too. This research paper discusses and analyzes of the trends associated with data mining and application involved in resume generalization through support vector machine i.e.SVM and term frequency-inverse document frequency i.e. TF-IDF 600 resumes are categorized with the SVM and TF-IDF models wherein 200resumes are for the testing set and the 400 resumes are for training set and their accuracy is determined with respect to the categorization based on the skill set. The process of getting the desired outcomes from the SVM and TF-IDF models we have implemented a method from the areas of finding out the document from which the desired data can be present, then the information retrieval process is done, then for human intelligence NLP (Natural Language Processing) is applied on it, then the required information is extracted and data is called mined by then. All current areas act as a single unit as non can be omitted even if we have to apply the complex methods like TF-IDF and SVM they also work on the same methodology.