RFM analysis for detecting future core technology

Authors:
Dohyun Kim;June Young Lee;Sejung Ahn;Yeongho Moon;Oh-Jin Kwon
Affiliations:
KISTI, Hoegiro, Dongdaemun-gu, Seoul, Korea;KISTI, Hoegiro, Dongdaemun-gu, Seoul, Korea;KISTI, Hoegiro, Dongdaemun-gu, Seoul, Korea;KISTI, Hoegiro, Dongdaemun-gu, Seoul, Korea;KISTI, Hoegiro, Dongdaemun-gu, Seoul, Korea
Venue:
Proceedings of the 2012 ACM Research in Applied Computation Symposium
Year:
2012

Citing 5
Cited 0

Training Cost-Sensitive Neural Networks with Methods Addressing the Class Imbalance Problem

IEEE Transactions on Knowledge and Data Engineering
The class imbalance problem: A systematic study

Intelligent Data Analysis
Knowledge discovery on RFM model using Bernoulli sequence

Expert Systems with Applications: An International Journal
Learning from Imbalanced Data

IEEE Transactions on Knowledge and Data Engineering
The citation life cycle of articles published in 13 American Psychological Association journals: A 25-year longitudinal analysis

Journal of the American Society for Information Science and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

RFM is a simple and powerful method to provide a framework for understanding and quantifying customer behavior based on purchase in marketing field. The purpose of this study is to demonstrate that RFM analysis can be effectively used for predicting future core technologies. Experimental results obtained using the US patent data show that recency, frequency, and monetary are efficient variables to identify the future core patents. In addition, the rules to identify the future core technology are searched using the classification and regression tree (CART), combined with the two sampling methods (over- and under-sampling) and the learning algorithms are compared in terms of precision, recall, and F-measure. Computational studies demonstrate that over-sampling method is effective for finding rules from imbalanced data, such as the data for detecting future core technology.