Internship Diary- Chapter 5

AI/ML and Models

Data science works in conjunction with AI and ML to extract valuable insights and knowledge from data. Data scientists utilize various techniques and tools to collect, clean, analyze, and interpret data. AI, as a broader field, incorporates ML and deep learning to provide intelligent capabilities to data science. Here's how they fit into the broader field:

Machine Learning (ML): ML is a key component of data science. It provides algorithms and models that enable machines to learn from data and make predictions or decisions. ML enables predictions and classifications of new, unseen data by training models on historical data.

ML is like a student who learns from examples. The more examples the student studies, the better they recognize patterns, make accurate predictions, and are also able to solve similar problems in the future.

Deep Learning: Deep learning is a subset of ML that uses artificial neural networks with multiple layers (deep neural networks) to process and extract features from data. Deep learning has revolutionized several areas of data science, such as image and speech recognition, natural language processing, and recommendation systems.

Deep learning is like a complex network of interconnected workers. Each worker (neuron) performs a simple task, but when they work together in multiple layers, they can tackle complex problems.

AI Applications: AI encompasses a broader set of technologies and techniques aimed at creating intelligent systems that can perform tasks that typically require human intelligence. ML and deep learning are crucial components of AI, as they provide the learning and decision-making capabilities necessary for AI systems.

We can imagine AI as a general-purpose toolbox with various tools. Each tool represents a different technique or algorithm that can be used to build intelligent systems. The toolbox provides a wide range of capabilities for solving diverse problems.

I have witnessed the application of AI/ML models while working at Tiger. Each model offers unique functionality and is selected based on the specific characteristics of the data and the goals of the task at hand. Let’s understand a few-

Random Forest Classifier: The Random Forest classifier is a popular machine learning model used for both classification and regression tasks. It operates by constructing an ensemble of decision trees, where each tree is trained on a random subset of features and data samples. During prediction, the model combines the predictions of multiple decision trees to determine the final output. Random Forest is versatile, interpretable, and widely used due to its excellent performance and resistance to outliers.

Logistic Regression: Logistic Regression is a statistical model commonly used for binary classification problems. Despite its name, it is primarily used for classification rather than regression tasks. Logistic Regression works by assigning weights to features and using them to calculate the likelihood of a data point belonging to a specific class. Logistic Regression is computationally efficient, performs well with linearly separable data, and provides insights into the importance and direction of features.

Gradient Boosting Classifier: The Gradient Boosting Classifier is an ensemble learning method that combines multiple weak predictive models, typically decision trees, to create a strong predictive model. It iteratively builds models in a stage-wise manner, where each subsequent model corrects the mistakes made by the previous model. It improves the performance of weak models by focusing on the instances that are difficult to classify correctly. Gradient Boosting is known for its high predictive accuracy, robustness against overfitting, and flexibility to handle various types of data.

These models, along with other techniques in data science, empower organizations to extract insights, make data-driven decisions, and solve complex problems across a wide range of industries and domains.