Albanian advertising keyword generation and expansion via hidden semantic relations

Abstract

Keyword generation and expansion are important problems in computational advertising. Keyword suggestion methods help advertisers to find more appropriate keywords. They involve discovering new words or phrases related to the existing keywords. Producing the proper hidden yet semantically relevant keywords is a hard problem. The problems real difficulty is in finding many such words. In this paper we propose an artificial keyword suggester for Albanian language by mimicking the human like systems. The possibility of a human to provide these keywords counts on the richness and deepness of its language and cultural qualifications. In order to provide additional keywords a human must accomplish multiple memory search tasks for meanings of huge number of concepts and their frame of references. Hence the memory of the proposed artificial keyword suggester is based on a large information repository formed by utilizing machine reading techniques for fact extraction from the web. As a memory we indirectly use the Albanian worldwide- web and the Gjirafa.com as a search engine. Complementary, the brain of the system is designed as a spreading activating network. The brain treats provided keywords and finds associations between them and concepts within its memory in order to incrementally compute and propose a new list of potential keywords. Experimental results show that our proposed method can successfully provide suggestion that meets the accuracy and coverage requirements.