Keyword extraction from social media via AHP

Abstract

In the context of natural-language processing, keyword extraction has been studied widely. In promoting business-enterprise goods and services, however, a major challenge remains to extracting keywords effectively and efficiently from social-media user-generated data, wherein employed are traditional, language-dependent and supervised keyword-extraction techniques. This study contributes a keyword extraction analytic hierarchy process (KEAHP), as a language-independent and unsupervised keyword-extraction technique. By using four user-generated data attributes, KEAHP identifies keywords from the word co-occurrence in linguistic networks, based on a multiple-attribute decision-making approach. The proposed technique has been validated via a publically-available standard dataset, and the experimental results show the effectiveness and efficiency of the algorithm in KEAHP. Despite its limitations, the study contends that KEAHP can drastically improve performance in promoting business-enterprise goods and services, while also discussed are implications for future research and practice in keyword-extraction techniques.