Automatic keyphrase extraction techniques: A review

Keyphrases are useful in organizing an overwhelming number of resources while providing ease of access for information retrieval. However, manually assigning keyphrases to a document is very expensive in terms of both human resources and time consumption. Therefore, there is a need for automatic keyphrase extraction. While techniques for automatic keyphrase extraction have been researched for a number of years, the accuracy rate remains low. This paper reviews existing automatic keyphrase extraction techniques with the goal of understanding their strengths and weaknesses. In future improvement of automatic keyphrase extraction technique, these identified strengths should be preserved while the weaknesses should be addressed so that a higher accuracy rate can be achieved. The results show that two important sub-processes should be the focus for further research. The paper also proposes a future study to address this issue.