TY - JOUR T1 - Evolution of Integrated Marketing Communication Research through Latent Dirichlet Allocation (LDA) Analysis JO - Journal of Emerging Trends in Marketing and Management VL - I IS - 1 SP - 61 EP - 70 PY - 2020 DA - 2020/08/31 PB - The Bucharest University of Economic Studies Publishing House PP - Bucharest, Romania T2 - AU - Popa, Alina AU - Brandabur, Raluca-Ecaterina SN - 2537-5865 DO - UR - http://www.etimm.ase.ro/RePEc/aes/jetimm/2020/ETIMM_V01_2020_48.pdf KW - Integrated Marketing Communication (IMC) KW - Latent Dirichlet Allocation (LDA) KW - Topic Modelling AB - Integrated Marketing Communication (IMC) is an area that emerged as a shift in the way MarCom departments were functioning at the beginning of 90's. For the last 30 years, the concept evolved from being a tactical set of actions to a customer-focused strategy. Despite the great interest in the field and empirical studies that showed the great impact of implementing the concept in organisations, there are no studies that would have extracted the tendencies in the whole field of IMC development of the last decade. The purpose of this study is to investigate the general research trends with an emphasis on what topics were mostly in focus, which ones were diminished in order to understand the life cycle of the IMC theory and practice. This study analysed the distribution of topics in each of the research papers from the IMC area published in the last 10 years using Latent Dirichlet allocation (LDA), an unsupervised topic modelling approach that extracts topics from a collection of documents. The results were then compared against other content analysis studies from the previous decade. Education in the IMC area, measurement and performance were found to be the topics of the greatest interest and growth. This denotes the transition of the domain from building the unanimously accepted theoretical basis to the practical part of the concept like efficient implementation, measurement and monitoring of the performance. This study answers the question regarding IMC development stage placing it into maturity and identifies a slight decline in the overall efforts in the area. Also, the paper enables researchers with an example methodology on how to use a machine learning approach for efficient, unbiased and replicable content analysis. Further studies are needed to understand how the topics in the IMC area evolved over time and how they relate to topics in related fields. ER -