Project Report

Creating a digital marketing campaign dashboard in tablea

A PROJECT REPORT

Submitted by

Khushi kapoor (21BCS2394)

in partial fulfilment of the award of the degree of

BACHELOR OF ENGINEERING

IN

COMPUTER SCIENCE ENGINEERING

Chandigarh University

Aug 2023
 BONAFIDE CERTIFICATE

Certified that this project report “Creating a digital campaign dashboard in tableau
is the bonafide work of the 'Khushi kapoor' who
carried out the project work under our supervision.

SIGNATURE SIGNATURE

(E13355) Natasha Sharma

HEAD OF THE DEPARTMENT SUPERVISOR

Submitted for the project viva-voce examination held on

INTERNAL EXAMINER EXTERNAL EXAMINER

 TABLE OF CONTENTS

List of Figures...........................................................................................................i
List of Tables .......................................................................................................... ii
Abstract.................................................................................................................. iii
Graphical Abstract ............................................................................................... iv
Abbrevations ...........................................................................................................v
Symbols .................................................................................................................. vi

CHAPTER 1. INTRODUCTION ........................................................................6

1.1. Identification of Client/ Need/ Relevant Contemporary issue ....................................6-7

1.2. Identification of Problem ............................................................................................... 7

1.3. Identification of Tasks ................................................................................................... 8

1.4. Timeline ......................................................................................................................... 9

1.5. Organization of the Report .......................................................................................... 10

Abstract:

In today's digital landscape, effective marketing

strategies are crucial for businesses to thrive. This
project focuses on creating a comprehensive Digital
Marketing Campaign Dashboard using Tableau, a
powerful data visualization tool. The dashboard aims
to provide marketers with insights into campaign
performance, audience engagement,
and return on investment (ROI) across various digital
channels such as social media, email, and web
advertising.

The dashboard integrates data from multiple sources,

including Google Analytics, social media platforms,
and email marketing tools, enabling a holistic view of
marketing efforts. Key performance indicators (KPIs)
such as click-through rates, conversion rates, and
customer acquisition costs are visualized through
interactive charts and graphs, allowing for real-time
analysis and data-driven decision-making
 CHAPTER 1.
INTRODUCTION

In Kidney diseases represent a formidable challenge in modern healthcare, affecting millions

of individuals worldwide and imposing a significant burden on healthcare systems. These
diseases, including chronic kidney disease (CKD), acute kidney injury (AKI), and diabetic
nephropathy, manifest in diverse forms, progress variably, and often remain asymptomatic
until advanced stages. Timely diagnosis and effective management are pivotal for preventing
disease progression, reducing complications, and improving patients' quality of life. In this
context, the application of machine learning (ML) techniques to predict kidney diseases
emerges as a promising avenue, offering the potential to enhance diagnostic accuracy,
streamline patient care, and alleviate the socioeconomic impact of these conditions.

1.1. Identification of Client /Need / Relevant Contemporary issue

• Identification of Client: The primary client for this research paper is the
healthcare industry, including medical practitioners, researchers, and
healthcare institutions, as well as patients and their families.
Additionally, policymakers and public health organizations interested in
improving kidney disease management and reducing healthcare costs are
potential stakeholders.

• Need: The pressing need addressed by this research paper is the early
detection and improved management of kidney diseases, which represent
a significant contemporary healthcare issue. Kidney diseases, such as
chronic kidney disease (CKD) and acute kidney injury (AKI), have seen
a rising prevalence globally due to factors like aging populations,
increased rates of diabetes and hypertension, and lifestyle-related risk
factors. Early diagnosis and intervention are crucial for preventing
kidney disease progression and its associated complications, including
end-stage renal disease (ESRD), which requires costly treatments like
dialysis or transplantation.

• Relevant Contemporary Issue: Kidney diseases have become a

growing public health concern, impacting both individuals and
healthcare systems. The contemporary issue lies in the inadequacy of
traditional diagnostic methods and the increasing burden of kidney
disease on healthcare resources. Early identification of kidney diseases
is vital to improving patient outcomes and reducing the economic strain
on healthcare systems. Moreover, the application of machine learning
(ML) and data-driven approaches in healthcare has gained significant
traction in recent years, making it a contemporary issue to explore and
develop ML-based solutions for kidney disease prediction.
 • This research paper addresses the timely and relevant issue of kidney
disease prediction using Python-based machine learning techniques. It
aims to provide valuable insights into the development and
implementation of ML models for early diagnosis, risk assessment, and
personalized healthcare management in the context of kidney diseases.
By addressing this contemporary need, the paper contributes to
advancing healthcare practices and reducing the societal and economic
burden associated with kidney diseases.

1.2. Identification of Problem

The problem addressed in this research paper is the challenge of accurately

predicting and diagnosing kidney diseases using machine learning
techniques in the context of healthcare. Kidney diseases encompass a range
of conditions, including chronic kidney disease (CKD), acute kidney injury
(AKI), diabetic nephropathy, and more, which can have severe
implications for patient health and well-being. The problem can be broken
down into several key components:

Complexity and Variability of Kidney Diseases: Kidney diseases are

characterized by their complexity and heterogeneity. They can have
various etiologies, progress at different rates, and exhibit diverse clinical
manifestations. Accurate prediction requires models that can capture this
complexity and adapt to different disease subtypes and stages.

Data Integration and Quality: Healthcare data relevant to kidney disease

prediction are often dispersed across various sources, including electronic
health records (EHRs), laboratory results, medical imaging, and genetic
information. Integrating and preprocessing these data to ensure their
quality and relevance pose significant challenges.

Class Imbalance: In many real-world healthcare datasets, there is a class

imbalance issue where the number of patients with kidney diseases
(positive class) is significantly smaller than those without kidney diseases
(negative class). This imbalance can lead to biased model predictions and
reduced sensitivity in detecting true cases.

Interpretability and Trustworthiness: The black-box nature of some

machine learning models, such as deep neural networks, can hinder their
adoption in clinical practice. Clinicians require interpretable models that
provide explanations for their predictions, ensuring trust and facilitating
informed decision-making.

Ethical Considerations and Bias: Ensuring fairness and avoiding bias in

kidney disease prediction models is crucial, as biased algorithms may
disproportionately affect certain demographic groups. Ethical
considerations, transparency, and equity in healthcare are important issues
that need to be addressed.
 Generalization and External Validation: Building prediction models
that generalize well across different patient populations and healthcare
settings is essential. Models should not be overly specialized to a single
dataset but should perform well in diverse and real-world scenarios.

Scalability and Computational Efficiency: Scalability is a concern when

implementing machine learning models in clinical settings. Models must
be computationally efficient to handle large datasets and make real-time
predictions.

Clinical Adoption: Bridging the gap between research and clinical

practice is a challenge. For machine learning-based kidney disease
prediction to be beneficial, it must be integrated into the existing healthcare
infrastructure and workflows, which may require changes in clinical
practices and policies.

Addressing these problems is essential to harness the full potential of

machine learning in predicting kidney diseases. The research paper aims to
explore and propose solutions to these challenges, ultimately contributing
to improved early diagnosis, patient care, and healthcare outcomes in the
context of kidney diseases.
.

1.3. Identification of Tasks

To have a Kidney diseases prediction, the following tasks may need to be completed:
1. Literature Review: Conduct a comprehensive review of existing literature related to
kidney disease prediction, machine learning techniques, and healthcare applications.
Type your text
Identify gaps, limitations, and emerging trends in the field.
2. Data Collection and Preprocessing: Gather relevant healthcare data sources,
including electronic health records, laboratory reports, imaging data, and genetic
information.
Preprocess and clean the data to handle missing values, outliers, and ensure data quality.
3. Model Selection: Analyze the dataset to identify relevant features for kidney disease
prediction.
Apply feature engineering techniques to create informative variables.
Address issues related to multicollinearity and dimensionality reduction.
4. Model Training and Evaluation: Evaluate various machine learning algorithms
suitable for kidney disease prediction.
 Experiment with algorithms such as decision trees, support vector machines, logistic
regression, random forests, neural networks, and ensemble methods.
Choose the most appropriate model(s) based on performance metrics.
5. Interpretability and Explainability: Split the dataset into training, validation, and test
sets.
Train the selected machine learning models on the training data.
Evaluate model performance using metrics like accuracy, precision, recall, F1-score,
ROC AUC, and calibration curves.
Address the class imbalance issue through techniques like oversampling or under-
sampling.
Interpretability and Explainability:
6. Bias and Fairness Assessment: Assess and mitigate biases in the model predictions to
ensure fairness and equity in healthcare outcomes.
Implement fairness-aware machine learning techniques to address potential disparities.
7. Validation and Generalization: Validate the model's performance on external datasets
to test its generalizability.
Investigate the model's robustness and adaptability to different healthcare settings and
patient populations.

1.4. Timeline

The following is a visual representation in the form of a Gantt chart that displays the
order in which tasks must be completed to finish the project. The chart is organized
by weeks and shows the start and end dates for each task:

Note: The dates on the chart correspond to the start date of the project.
Month 1-2: Project Initiation and Literature Review
Define project objectives and scope.
Conduct an extensive literature review on kidney disease prediction, machine
learning, and healthcare applications.
Identify relevant datasets and data sources for the research.

Month 3-4: Data Collection and Preprocessing

Acquire and preprocess healthcare data, ensuring data quality and consistency.
Perform initial data exploration to understand the dataset's characteristics.
Address missing values, outliers, and data cleaning.
 Month 5-6: Feature Selection and Engineering
Analyze the dataset to identify informative features for kidney disease prediction.
Apply feature engineering techniques to create new relevant variables.
Perform dimensionality reduction if needed to handle high-dimensional data.

Month 7-8: Model Selection and Training

Evaluate various machine learning algorithms suitable for kidney disease prediction.
Experiment with different model architectures, such as decision trees, support vector
machines, neural networks, and ensemble methods.
Train and validate models using appropriate evaluation metrics.

Month 9-10: Interpretability and Explainability

Ensure that the selected models provide interpretable and explainable predictions.
Generate feature importance rankings and explanations for model predictions.
Implement techniques to make model outputs accessible to healthcare professionals.

Month 11-12: Bias Mitigation and Fairness Assessment

Assess and mitigate biases in the model predictions to ensure fairness and equity.
Implement fairness-aware machine learning techniques to address potential
disparities.

Month 13-14: Validation and Generalization

Validate model performance on external datasets to test generalizability.
Investigate the model's robustness and adaptability to different healthcare settings
and patient populations.

Month 15-16: Ethical Considerations and Approvals

Address ethical considerations in the use of patient data for research.
Ensure compliance with data privacy regulations and obtain necessary approvals
from ethics committees or relevant authorities.

Month 17-18: Results Interpretation and Discussion

Interpret the research results and discuss their implications.
Relate the model's performance to clinical relevance and patient outcomes.
Identify limitations and challenges encountered during the research.

Month 19-20: Future Directions and Recommendations

Suggest potential avenues for future research in kidney disease prediction.
Provide recommendations for healthcare practitioners, researchers, and
policymakers based on the study's outcomes.

1.5. Organization of the Report

Chapter 1: Introduction

 This chapter includes the identification of the clients as well as the project's purpose.
The challenges of developing an AI-based drone are also discussed, as well as the
tasks required to build a prototype and overcome these challenges.

Chapter 2: Literature Survey

 In this section of the report, we examine the prior or current endeavours made by
other developers in the drone industry. Furthermore, we analyse the advantages and
disadvantages of their model and compare it with ours.

Chapter 3: Data Collection and Preprocessing:

 Explanation of data sources and acquisition methods.
 Data preprocessing steps, including cleaning, handling missing values, and ensuring
data quality.
 Initial data exploration and visualization.

Chapter 4: Identification of relevant features for kidney disease prediction.

 Description of feature engineering techniques applied.
 Dimensionality reduction methods used, if applicable.

 This chapter focuses on presenting the outcomes and results achieved from the
project, which includes testing and implementing different algorithms and hardware
components utilized in the development of the prototype.

Chapter 5: Result

Real-Time Performance Monitoring: Marketing teams will be able to track their

campaigns in real-time, making it easier to optimize underperforming ads.
Campaign Comparison: The dashboard will allow users to compare the
performance of different campaigns, platforms, and traffic sources.
 Chapter 6: Conclusion and Future scope

•The Tableau digital marketing campaign dashboard is an essential tool for monitoring.
the sucess of marketing efforts.By providing real-time insights into key metrics, this
dashboard empowers marketers to make data-driven decisions and optimize their
Campaigns effectively.