Identification of Telecom Volatile Customers Using a Particle Swarm Optimized K-Means Clustering on Their Personality Traits Analysis

Abstract

This research uses the telecom customers personality traits (extraversion, agreeableness, and neuroticism) to identify the volatile customers that always use the negative word of mouth (NWOM) in communications with others. Hence, a combination of text analysis and a personality analysis tool has been used to determine the customers personality factors from their chatting textual data, A particle swarm optimized k-means was used in the clustering process. The results provide an overview on how a chatbot conversation text represent the customer behavior. Optimizing the k-means cluster using partial swarm achieves a higher accuracy than using the traditional clustering technique.

Customer Loyalty to Modern Retail Channel A Study in The Context of Covid-19 Pandemic in Vietnam

“…The other two factors of the model are the personality traits (H4) and commitment (H5) are further explored as factors affecting customer loyalty to a lower level. The results of this study (H4) are supported by recent research (Elfergany & Adl, 2020). An important finding of this study is that customer commitment weakly affects loyalty when placed in the context of Covid -19 pandemic.…”

“…Studies of consumer behavior have shown that characteristics can help develop the integrated conceptual frameworks for understanding the consumers and enable the development of better targeted communications (Baumgartner, 2002) as well as the most influential and relevant characteristics of consumers' conversion behavior (De Wulf et al 2001). When a customer appreciates a certain brand and makes repetitive purchase from that brand due to the favorable traits that suit the customer's characteristics, he or she will have a positive image of that brand and is more likely to become a loyal customer (Kim et al 2018;Elfergany & Adl, 2020;. Previous studies have shown a strong relationship between customer loyalty and characteristics.…”

Indonesian pharmacy retailer segmentation using recency frequency monetary-location model and ant K-means algorithm

“…Those data can be used to cluster retailers based on similar buying behavior. Combined with swarm intelligence algorithm such as ant Clustering, the retailer segmentation cluster can be generated [13]- [16]. Market segmentation is an approach to cluster market player with like-characteristics given the market's heterogenistic nature [17].…”

An Adaptive Neural-Fuzzy Inference System for Prediction of Muscle Strength of Farmers in India

“…Dif and Elberrichi (2020) have used "inception-v3 convolutional neural-network architecture", six histopathological-source datasets, and four target-sets as base-modules and revealed the importance of the pre-trained histopathological-models compared to the ImageNet-model. The use of particle swarm optimized k-means cluster help in achieving a higher-accuracy in comparison to the traditional clustering-technique (Elfergany and Adl, 2020). However, the basis of ANFIS is driven from the approach of fuzzy modeling and the fuzzy inference system (FIS) permits extracting the model from the data of input or output (Zadeh, 1973;Krueger et al 2011).…”