A Novel Metadata Based Multi-Label Document Classification Technique

From the beginning, process of research and its publication is an ever-growing phenomenon with emergence web technologies, growth rate overwhelming. On a rough estimate, more than thirty thousand journals have been issuing around four million papers annually on average. Search engines, indexing services, digital libraries searching for such publications over web. Nevertheless, getting most relevant articles against user requests yet fantasy. It mainly because are not appropriately indexed based hierarchies granular subject classification. To overcome this issue, researchers striving to investigate new techniques classification especially, when complete article text available (a case non-open access articles). The proposed study aims multilabel metadata in best possible way assess, “to what extent metadata-based features can perform contrast content-based approaches.” In regard, novel investigating proposed, developed, evaluated as Title Keywords articles. technique has assessed two diverse datasets, namely, from Journal universal computer science (J.UCS) benchmark dataset comprises published by Association computing machinery (ACM). yields encouraging results state-of-the-art literature.