The telecom industry is saturated with many service providers competing for highly rational customers. The current big data and highly technological era calls for real-time churn analysis and decision making which has also been highlighted in previous studies. However, telecom data is highly dimensional in nature thus when this is coupled with this big data era increases the computational and processing costs. Therefore, this complexity and dimensionality of telecom data coupled with the current need for near or real-time churn analysis demands feature selection-based models that efficiently consider the most relevant variables in explaining customer churn behaviors. This study proposes a feature extraction-based churn prediction model that concentrates on the most relevant features with significant discriminatory power for churn. The data has been reduced on the basis of missing values and irrelevant variables. Irrelevant variables were first identified by use of Random Forest and Logistic Regression models. The findings of the study provide churn analysts with insights about the prediction errors to consider and minimize in their future churn analyses. It also contributes to reducing computational costs incurred by churn analysts working with big data in their churn prediction and analysis.
Microblogging services, especially Twitter, allow the user to share their most recent thoughts, feelings or news freely and almost immediately. Hence, the number of news tweets generated by the news media is increasing exponentially. Mining the valuable data from the large volume of tweets can help increase the revenue of organisations by allowing them to engage with the public faster and better by responding to the latest topics of interest. In this work, mining the hot keywords and being able in classifying the news tweets, trending topic and keywords in the news tweets. Both supervised and unsupervised machine learning models are used. Several machine learning algorithms are being used to compare the accuracy in classifying the tweets.
Customer churn which also commonly referred to as customer attrition, occurs when a customer ceases or stops doing business with a particular company or service. Predicting customer churn had become one of the major aim of businesses is various sectors, namely the telecommunication sector as the markets are very saturated. Apart from that, the cost to retain existing customers is much lesser as compared to the cost to acquire or attract new customers through marketing. This paper proposes a new hybrid algorithm which incorporate the algorithms of Particle Swarm Optimization (PSO) as well as Extreme Learning Machine (ELM) to build a telecommunication churn prediction model which can accurately predict churners and non-churners. This model will be named as Particle Swarm Extreme Learning Machine (PSELM) model. PSO algorithm is able to effectively scale the selected features so that ELM algorithm can classify data based on these features more easily and provide accurate classification.
Customer insighs is the key to the success of e-commerce. Therefore, factors affecting customer satisfaction leading to product purchase and re-purchase should be studied extensively. This study intends to identify the key drivers that influence the satisfaction and the model which can predict the likelihood of customer satisfaction. The outcome would provide insights to prioritise factors that are significant, as well as to provide advice to a wide range of sellers. Four classification machine learning algorithms decision tree, random forest, artificial neural network and support vector machine are evaluated to classify customer satisfaction based on a 3-year historical data from an e-commerce retailer. There were a few challenges with the dataset, such as imbalanced, skewed and missing. Data pre-processing was conducted, and different techniques were evaluated. Of the algorithms evaluated, the best result is achieved by Random Forest with the highest accuracy and reasonable processing time. Meeting the estimated delivery date and the number of days taken to deliver an order is found to be the top two important factors affecting customer satisfaction.
Rapid expansion of IoT technologies and their devices has created the considerable convenience in the daily lives of the people. Modernization of technique in IoT has allowed them to perform just more than sensing that directly means the more energy consumption. It is a known fact that most of the IoT devices have the limited power supply and hence it is an obvious need to design the energy efficient model. Despite of IoT devices being so useful in daily life and in other industries, data security is one of the primary concern in the fields of IoT application such as healthcare, agriculture, defense etc., and it is a challenging task to provide the data security in these fields. In past, several methods have been proposed to address this challenge, however, either they failed to provide the security or they lack from the efficiency. In this paper, it introduces a methodology named as ESDAM (Efficient and Secure Data Aggregation against Malicious Nodes) that provides better security with improved efficiency. The proposed methodology is parted into two methods, which helps in discerning the malicious nodes. First method, is through extending coordinates and the second method, is through surveilling the adjacent nodes. Extensive simulation has been performed by applying various constraints through persuading various number of malicious and performance metrics such as energy utilization, number of failed nodes, packet mismatch rate and packet discern rate. The performance evaluation of the simulation results proves that proposed methodology performs better than the existing methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.