DSpace JSPUI
DSpace preserves and enables easy and open access to all types of digital content including text, images, moving images, mpegs and data sets
Learn More
Please use this identifier to cite or link to this item:
http://dspace.dtu.ac.in:8080/jspui/handle/repository/17995Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Sharma, Sadgi | - |
| dc.date.accessioned | 2020-08-26T07:50:34Z | - |
| dc.date.available | 2020-08-26T07:50:34Z | - |
| dc.date.issued | 2020-07-20 | - |
| dc.identifier.uri | http://dspace.dtu.ac.in:8080/jspui/handle/repository/17995 | - |
| dc.description | Business Analytics is the thorough analysis of technologies, skills, practices and continuous evaluation of the business performance to come up with new improvements and provide insights on how to proceed resourcefully. Business analysis attempts to interpret the business performance based on data and statistical methods. The future decisions regarding the businesses come after a successful business analysis. Business analytics can be sub-divided into two components namely business intelligence and statistical analysis. Both are the underlying concepts that are used to determine customer churn for an online business. 1. Business Intelligence covers the research on past data to evaluate the performance score of the management team, staff, and the business department. This helps the startups/established companies to relate the efforts of the company with its outcomes. This further carries out the area of focus and improvement. It does not include heavy predictive modelling or forecasting. 2. Statistical Analysis involves the use of predictive analysis using statistical methods. It is useful in figuring out the future performance of the product. Customer Churn Analysis also uses statistical analysis techniques based on cluster analysis in order to develop group of customers on the basis of similar characteristics so that the marketing campaigns can be aimed at achieving set goals. Customer Churn is also known as Customer Attrition and refers to a situation when a client or a subscriber ceases his or her affiliation with the business. Customer may vary from industry to industry for example for a Gaming Industry, customer is a player, for a Telecom Industry, customer is a subscriber, for a Product company, it may be a user. It has been found that in most Online Businesses, typically a customer drops using a service once a considerable amount of time has passed. This costs company and it becomes important to replace the churned customer by acquiring new customers. Minimizing the customer churn is the key business goal of every company specially a company that mostly does its business online. The cost incurred to get the customer on board initially is also booked as a loss in revenue, once the company loses the customer later. Furthermore, acquiring a customer is rather more difficult and expensive than it is to retain a paying customer. Since the boom in online business, this has become a major problem that needs a solution immediately. Here, in this project the data considered for Analysis is based on IBM Watson Telco Customer Churn dataset which is retrieved on 15th April 2020. The report discusses using Machine Learning and Neural Network Techniques along with Power BI Visualization to understand and find the customers that have 8 9 high probability of churning in near future. It discusses one of the major market drivers of revenue. Customers are considered to be the fuel that controls a business. A loss in customer will impact sales for a company. Furthermore, as mentioned earlier it is much more costly and time consuming acquire new customers in opposition to holding the existing ones. Thus, the need of the hour for organizations is to emphasize on plummeting customer churn. | en_US |
| dc.description.abstract | With the increase in competition and technological advancements in the Telecommunications Industry, being able to assess and predict the customer attribution ahead of time is of extreme importance to a company. This Research project describes the application of Machine Learning and Neural Network techniques in R and Python to predict customer churn. Customer churn is considered to be the major problem in the telecommunications industry. Numerous studies have shown that attracting new customers in a telecommunications industry is much more expensive than retaining existing ones. Therefore, companies are focusing on developing accurate and reliable predictive models to identify potential customers that will churn in the near future. The aim of this project is investigating the main reasons for churn telecommunications sector. This major project report provides a comparative study on different machine learning techniques used for predicting customer churn. The study also covers different phases including business scenario, data analysis, data pre-processing, and implementation of different algorithms for classification. The resulted models have outperformed the evaluation metrics of different research papers published in the similar domain. For the businesses that offer subscription-based services, it is important to predict customer churn as well as elucidate the parameters related to attrition. The predictive analytics-based techniques like logistic regression can be not as much of precise than new techniques such as machine learning and ensemble models. This report discusses using Artificial Neural Network and Ensemble Models in R and Python to predict the customer churn. | en_US |
| dc.language.iso | en | en_US |
| dc.subject | Machine Learning | en_US |
| dc.subject | Algorithms | en_US |
| dc.subject | Neural Network techniques | en_US |
| dc.subject | Business Analytics | en_US |
| dc.title | A Comparative Study on Machine Learning Algorithms for Customer Churn Analytics with Power BI Dashboard | en_US |
| dc.type | Other | en_US |
| Appears in Collections: | MBA | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Comparative Study on Machine Learning.pdf | Business Analytics is the thorough analysis of technologies, skills, practices and continuous evaluation of the business performance to come up with new improvements and provide insights on how to proceed resourcefully. Business analysis attempts to interpret the business performance based on data and statistical methods. The future decisions regarding the businesses come after a successful business analysis. Business analytics can be sub-divided into two components namely business intelligence and statistical analysis. Both are the underlying concepts that are used to determine customer churn for an online business. 1. Business Intelligence covers the research on past data to evaluate the performance score of the management team, staff, and the business department. This helps the startups/established companies to relate the efforts of the company with its outcomes. This further carries out the area of focus and improvement. It does not include heavy predictive modelling or forecasting. 2. Statistical Analysis involves the use of predictive analysis using statistical methods. It is useful in figuring out the future performance of the product. Customer Churn Analysis also uses statistical analysis techniques based on cluster analysis in order to develop group of customers on the basis of similar characteristics so that the marketing campaigns can be aimed at achieving set goals. Customer Churn is also known as Customer Attrition and refers to a situation when a client or a subscriber ceases his or her affiliation with the business. Customer may vary from industry to industry for example for a Gaming Industry, customer is a player, for a Telecom Industry, customer is a subscriber, for a Product company, it may be a user. It has been found that in most Online Businesses, typically a customer drops using a service once a considerable amount of time has passed. This costs company and it becomes important to replace the churned customer by acquiring new customers. Minimizing the customer churn is the key business goal of every company specially a company that mostly does its business online. The cost incurred to get the customer on board initially is also booked as a loss in revenue, once the company loses the customer later. Furthermore, acquiring a customer is rather more difficult and expensive than it is to retain a paying customer. Since the boom in online business, this has become a major problem that needs a solution immediately. Here, in this project the data considered for Analysis is based on IBM Watson Telco Customer Churn dataset which is retrieved on 15th April 2020. The report discusses using Machine Learning and Neural Network Techniques along with Power BI Visualization to understand and find the customers that have 8 9 high probability of churning in near future. It discusses one of the major market drivers of revenue. Customers are considered to be the fuel that controls a business. A loss in customer will impact sales for a company. Furthermore, as mentioned earlier it is much more costly and time consuming acquire new customers in opposition to holding the existing ones. Thus, the need of the hour for organizations is to emphasize on plummeting customer churn. | 1.61 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.