My journey migrating to ML(Machine learning) (in progress…)

I'm a +9y Fullstack engineer ,3 years ago I just decided to migrate to DataScience world, but why? what do you wanna do with this ? this was the question that not was too clear 3.4years ago , after ran into some well complex challenge in my work finally started realize where when I should consider user my in progress Datascience takeaways

Before start in machine learning , start learing these concepts :

Data Science:

• Data Science is an interdisciplinary field that combines statistics, mathematics, programming, and domain knowledge to extract insights and knowledge from structured and unstructured data.
• It encompasses the entire data lifecycle, from data collection and cleaning to data analysis and visualization.
• Data Science:
• The primary goal is to derive actionable insights from data, help in decision-making, and inform business strategies.
• Data Scientists often use various statistical and analytical methods to understand data patterns and trends.

Machine Learning:

• Machine Learning is a subset of artificial intelligence (AI) that focuses on developing algorithms and statistical models that enable computers to learn from and make predictions or decisions based on data.
• It primarily deals with the creation of models that can generalize from training data to unseen data.
• The goal of Machine Learning is to create predictive models that can automate tasks, classify data, and identify patterns without explicit programming for each specific task.
• ML focuses on improving the model’s accuracy over time through learning from new data.
• The primary goal is to derive actionable insights from data, help in decision-making, and inform business strategies.
• Data Scientists often use various statistical and analytical methods to understand data patterns and trends.

This is extremely important to know the data lifecycle

every data passes through this process before we get the final outcome.

• data collection: Gathering data from various sources (databases, APIs, web scraping).
• data cleaning: Preprocessing data to handle missing values, outliers, and inconsistencies.
• data analysis and exploratory data analysis (EDA): Analyzing data to find patterns, trends, and anomalies using statistical tools and visualizations.
• Modeling: Applying statistical and ML models to make predictions or classify data.
• data visualization and communication: Presenting findings through visualizations, reports, or dashboards to stakeholders.

Illustration of how ML simplifies complex predictions compared to traditional methods!

Traditional Approach:

• This approach relies on physical laws and mathematical equations to simulate the atmosphere. Scientists create detailed models of the Earth’s climate system, which involve complex fluid dynamics and thermodynamics equations to describe how air, water, and energy move around.
• To make accurate predictions, the model needs to consider countless factors like temperature, pressure, humidity, wind patterns, and interactions between these elements, making it highly complex and resource-intensive.

Machine Learning Approach:

• With ML, rather than coding in the physics or laws of the environment, we can feed the model historical weather data — temperature, pressure, wind speeds, and rainfall records over years.
• The ML model learns patterns from this data, recognizing how different weather conditions have led to varying rainfall levels in the past. Once trained, the model can predict rainfall by finding similar patterns in the current weather data.
• This approach simplifies the prediction process, as the model automatically derives the mathematical relationships rather than having them explicitly defined by scientists.

Categories of Machine Learning (ML) systems

Each category defined by how they learn and the type of task they are suited for.