What is Data Science

What is Data Science?

Data science is a multidisciplinary field that uses scientific methods, statistics, programming, and domain expertise to extract knowledge and actionable insights from data.

If data analytics answers "What happened?" and "Why?", data science goes further to answer "What will happen?" and "How can we make it happen automatically?"

Data scientists build models that can:

• Predict whether a loan applicant will default
• Recommend what product a customer should buy next
• Detect fraudulent transactions in real time
• Translate text between languages
• Classify medical images to detect diseases

---

The Three Pillars of Data Science

Data science sits at the intersection of three domains:

1. Mathematics and Statistics

The foundation of every model. Key areas include:

• Probability: How likely is an event?
• Statistics: How do we describe and compare data?
• Linear algebra: How do we represent and transform data mathematically?
• Calculus: How do we optimise models to perform better?

You do not need a mathematics degree to start. But understanding the intuition behind statistical concepts is essential.

2. Computer Science and Programming

Data scientists write code to:

• Collect and clean data
• Build and train models
• Deploy models to production systems

Python is by far the dominant language in data science, though R is also used in academic and research settings.

3. Domain Expertise

A data science model is only useful if it solves a real problem. Domain expertise helps you:

• Ask the right questions
• Understand whether a result makes sense in context
• Communicate findings to non-technical stakeholders

A data scientist working in healthcare who understands medicine will build far more useful models than one who does not.

---

What Data Scientists Actually Build

Machine Learning Models

Machine learning (ML) is a subset of artificial intelligence where systems learn from data rather than being explicitly programmed with rules.

Types of machine learning:

• Supervised learning: The model learns from labelled examples. You give it 10,000 emails labelled "spam" or "not spam" and it learns to classify new emails.
• Unsupervised learning: The model finds patterns in unlabelled data. It might discover that your customers cluster into five distinct groups based on their behaviour.
• Reinforcement learning: The model learns by trial and error, receiving rewards for good decisions. Used in robotics and game-playing AI.

Statistical Models

Not all data science is machine learning. Statistical models like linear regression, logistic regression, and survival analysis are powerful, interpretable, and widely used.

Natural Language Processing (NLP)

Teaching computers to understand and generate human language. Powers chatbots, sentiment analysis, machine translation, and document summarisation.

Computer Vision

Teaching computers to understand images and video. Powers facial recognition, medical imaging, self-driving cars, and quality inspection in manufacturing.

---

The Data Science Workflow

A data science project typically follows these stages:

1. Problem definition: What is the business problem? What would a useful prediction or insight look like?
2. Data collection: Gather relevant data from databases, APIs, web scraping, or sensors.
3. Exploratory Data Analysis (EDA): Understand the data's structure, quality, and patterns.
4. Data preparation: Clean, transform, and engineer features from the raw data.
5. Modelling: Select, train, and evaluate candidate models.
6. Evaluation: Assess model performance using appropriate metrics.
7. Deployment: Put the model into production where it can make real predictions.
8. Monitoring: Track model performance over time and retrain when it degrades.

---

Who Hires Data Scientists?

Data science is a global field, and Africa's growing tech sector is creating significant demand:

• Fintech companies: Flutterwave, Paystack, Kuda, and Moniepoint use data science for fraud detection and credit scoring.
• Telecommunications: MTN, Airtel, and Glo use it for churn prediction and network optimisation.
• E-commerce: Jumia and Konga use recommendation systems and demand forecasting.
• Healthcare organisations: Predict disease outbreaks and optimise resource allocation.
• Banks: Use credit risk models to decide who qualifies for loans.
• Global tech companies: Google, Meta, Amazon, and Microsoft all have large data science teams.

---

Is Data Science Right for You?

Data science is a great fit if you:

• Are comfortable with mathematics and enjoy problem-solving
• Like programming and building things with code
• Are curious and enjoy asking "why" and "what if"
• Can communicate complex ideas clearly

It requires more mathematical depth than data analytics, but the investment pays off in the ability to build systems that learn and improve automatically.