Table of ContentsIntroductionResearch and Investigation of TechniquesTechniques UsedB. Logistics RegressionConclusionCustomer satisfaction has a huge impact on the service delivery of any company. A simple word of mouth review structures the business environment to improve productivity and performance. With such impact from customers, it is essential to keep them on track, knowing the value of the product and service. The approach used for this project is to analyze digital media users, to check if they could continue business with the organization, if not, encourage them to do business with the help of greater delivery of services. For this analysis, a sample of data from digital media users was taken into account, in order to find out whether they could possibly be unsubscribed in the future. This prediction was made using machine learning techniques. The tool used for this analysis was Rapidminer. The result was presented with accurate results in statistical representation. Say no to plagiarism. Get a tailor-made essay on “Why Violent Video Games Should Not Be Banned”? Get Original EssayIntroductionIn general, CRM (Customer Relationship Management) is a tool that helps the organization to maintain the relationship between buyers and customer interaction, track their records and accounts. This helps them improve customer satisfaction. For analysis, a sample of digital media data was considered for churn prediction. This analysis aims to predict whether a customer would choose to stay with the organization even after the contract term. This is similar to the attrition model. Customer loyalty is an important aspect in any organization, where it shows the performance level of the company from the bottom to the top. Attrition is also one of the major jobs in data mining. In today's era, everything is becoming digital. The use of digital media is becoming a necessity for survival in the business environment. This helps organizations and customers stay abreast of the trend for their own needs. There are many forms of digital media in various formats such as audio, video, images and graphical representations. Considering the attrition model, there are three types, namely voluntary attrition, involuntary attrition and expected attrition. If a client wishes to change companies, this is a voluntary departure. Involuntary attrition, also known as forced attrition, occurs when the customer is terminated by the company for any reason, some common reasons being unpaid invoices. Expected attrition occurs when the customer is no longer available in the target area, such as when a customer moves to another location. There are several methods to predict the outcome of this project. The main background of this project is to focus on survival analysis. In this analysis, machine learning techniques are used to check the variation between them. These are deep learning and logistic regression. Through such techniques, the most accurate method will be known and can be taken into consideration. To perform this analysis, a tool called “Rapidminer” was used.Research and Investigation of TechniquesThere are various techniques available to implement and obtain results from the prediction of customer churn analysis in digital media. The techniques may be a machine learning technique such as aBayesian network, deep learning or decision trees. In another way, it can also be a statistical method of prediction via logistic regression, which is mainly carried out between a dependent variable and another variable when the dependent variable is dichotomous. Some previous work has been done on this project with certain techniques. All these techniques only gave the expected result. The dataset used for this project is very balanced. This helps ML techniques to perform analysis and give effective results. If there is an imbalance, the techniques will not work and effective results will not be available. However, for imbalanced datasets, there is a technique called oversampling technique, which deals with classification problems and has two types. These are the synthetic minority oversampling technique and the adaptive synthetic sampling technique. This technique helps balance data sets, making it easier to perform analysis. Another popular technique used for churn analysis is CART, which is a classification and regression tree model. This is the branch of the decision tree model. This technique mainly deals with classification and misclassification issues in the dataset. The other popular model used for this analysis was the Support Vector Machine (SVM) model. This model also works mainly on classification linearity problems. It is effective for working on linear and non-linear cases. The models mentioned above are not limited, but are worth mentioning when using them for this churn analysis. It has a special way of applying certain assumptions to be more effective. Techniques Used As noted previously, many important techniques are available. But in this project, only two techniques are used to find churn analysis in digital media. These techniques are very popular and widely used for this type of churn analysis project. This technique not only helps us predict the outcome, but also helps us statistically with all the factors that cause a customer to stay or choose another network. The dataset used for this project has 21 columns. The “Churn” column is the dependent variable. It is a dichotomous variable with yes or no. The independent variables are age, gender, months of service, telephone service, multiple lines, Internet service, online security, online backup, device protection, technical support, streaming TV and movies, contract length, paperless billing, payment method, monthly fees and Total Charges.A. Neural Networks (Deep Learning) This is one of the most popular algorithms in the field of prediction analysis. It is one of the branches of machine learning techniques. This Big Data processing is capable of analyzing a large amount of data at a given time, but it may also take a while to run the entire dataset if the data volume is very high. This technique is more flexible and scalable. The analysis was carried out using the Rapidminer tool. In this test, the accuracy is calculated with the global variables. The metric type for this test is binominal. Confusion matrix algorithm is used for statistical classification of the dataset. Using simulation, an in-depth understanding is analyzed of the type of customer who prefers convenience with bills..