Discovery of Long Tail Keywords in Paid SearchTesiero J*
Principal Data Scientist Consultant, University of Maine, USA
- Corresponding Author:
- Tesiero J
Principal Data Scientist Consultant
University of Maine, USA
Tel: 207 581 1865
Received date April 22, 2015; Accepted date July 25, 2016; Published date July 29, 2016
Citation: Tesiero J (2016) Discovery of Long Tail Keywords in Paid Search. J Appl Computat Math 5:315. doi: 10.4172/2168-9679.1000315
Copyright: © 2016 Tesiero J. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
The following work describes an elegant, efficient keyword clustering method to discover long tail keywords in paid search data. In keyword auctions, such words often go undiscovered as their cost in being bid to higher ranking positions is deemed too high to justify the potential of significantly added conversion revenue. By discovering clusters with low volume keywords and established, high-performing and high volume keywords, the quality of the low volume (long tail) keywords is inferred by association.
After a brief introduction, the data used to train the clustering algorithm is described. Then, the data reduction process (the discovery of the most predictive features) is described. We then describe the method, followed by the results and interpretation.