Log In
Or create an account ->
Imperial Library
Home
About
News
Upload
Forum
Help
Login/SignUp
Index
Title Page
Copyright and Credits
Managing Data Science
Dedication
About Packt
Why subscribe?
Contributors
About the author
About the reviewers
Packt is searching for authors like you
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the color images
Conventions used
Get in touch
Reviews
Section 1: What is Data Science?
What You Can Do with Data Science
Defining AI
Defining data science
The influence of data science
Limitations of data science
Introduction to machine learning
Decisions and insights provided by a machine learning model
Data for machine learning models
Origins of machine learning
Anatomy of machine learning
Main types of tasks you can solve with machine learning
Introduction to deep learning
Diving into natural language processing
Exploring computer vision
Deep learning use case
Introduction to causal inference
Summary
Testing Your Models
Offline model testing
Understanding model errors
Decomposing errors
Understanding overfitting
Using technical metrics
More about imbalanced classes
Applying business metrics
Online model testing
Online data testing
Summary
Understanding AI
Understanding mathematical optimization
Thinking with statistics
Frequentist probabilities
Conditional probabilities
Dependent and independent events
Bayesian view on probability
Distributions
Calculating statistics from data samples
Statistical modeling
How do machines learn?
Exploring machine learning
Defining goals of machine learning
Using a life cycle to build machine learning models
Linear models
Classification and regression trees
Ensemble models
Tree-based ensembles
Clustering models
Exploring deep learning
Building neural networks
Introduction to computer vision
Introduction to natural language processing
Summary
Section 2: Building and Sustaining a Team
An Ideal Data Science Team
Defining data science team roles
Exploring data science team roles and their responsibilities
Case study 1 – Applying machine learning to prevent fraud in banks
Case study 2 – Finding a home for machine learning in a retail company
Key skills of a data scientist
Key skills of a data engineer
Key skills of a data science manager
Getting help from the development team
Summary
Conducting Data Science Interviews
Common flaws of technical interviews
Searching for candidates you don't need
Discovering the purpose of the interview process
Introducing values and ethics into the interview
Designing good interviews
Designing test assignments
Interviewing for different data science roles
General guidance
Interviewing data scientists
Interviewing data engineers
Summary
Building Your Data Science Team
Achieving team Zen
Leadership and people management
Leading by example
Using situational leadership
Defining tasks in a clear way
Developing empathy
Facilitating a growth mindset
Growing the expertise of your team as a whole
Applying continuous learning for personal growth
Giving more opportunities for learning
Helping employees to grow with performance reviews
Case study—creating a data science department
Summary
Section 3: Managing Various Data Science Projects
Managing Innovation
Understanding innovations
Why do big organizations fail so often?
Game of markets
Creating new markets
Exploring innovation management
Case study – following the innovation cycle at MedVision
Integrating innovations
Balancing sales, marketing, team leadership, and technology
Managing innovations in a big company
Case study – bringing data science to a retail business
Managing innovations in a start-up company
Finding project ideas
Finding ideas in business processes
Finding ideas in data
Case study – finding data science project ideas in an insurance company
Summary
Managing Data Science Projects
Understanding data science project failure
Understanding data science management approaches
Exploring the data science project life cycle
Business understanding
Data understanding
Data preparation
Optimizing data preparation
Modeling
Evaluation
Deployment
Choosing a project management methodology
Waterfall
Agile
Kanban
Scrum
Choosing a methodology that suits your project
Creating disruptive innovation
Providing a tested solution
Developing a custom project for a customer
Estimating data science projects
Learning to make time and cost estimates
Discovering the goals of the estimation process
Summary
Common Pitfalls of Data Science Projects
Avoiding the common risks of data science projects
Approaching research projects
Dealing with prototypes and MVP projects
Case study – creating an MVP in a consulting company
Mitigating risks in production-oriented data science systems
Case study – bringing a sales forecasting system into production
Summary
Creating Products and Improving Reusability
Thinking of projects as products
Determining the stage of your project
Case study – building a service desk routing system
Improving reusability
Seeking and building products
Privacy concerns
Summary
Section 4: Creating a Development Infrastructure
Implementing ModelOps
Understanding ModelOps
Looking into DevOps
Exploring the special needs of data science project infrastructure
The data science delivery pipeline
Managing code versions and quality
Storing data along with the code
Tracking and versioning data
Storing data in practice
Managing environments
Tracking experiments
The importance of automated testing
Packaging code
Continuous model training
Case study – building ModelOps for a predictive maintenance system
A power pack for your projects
Summary
Building Your Technology Stack
Defining the elements of a technology stack
Choosing between core- and project-specific technologies
Comparing tools and products
Case study – forecasting demand for a logistics company
Summary
Conclusion
Advancing your knowledge
Summary
Other Books You May Enjoy
Leave a review - let other readers know what you think
← Prev
Back
Next →
← Prev
Back
Next →