When one encounters the term “Machine Learning,” it may initially seem grandiose or larger than life—a concept straight out of a sci-fi movie where programmers are hacking into systems or crafting artificial intelligence, fingers flying over keyboards typing a hundred lines of code per minutes. Personally, I hesitated to delve into Machine Learning, mainly because my initial exposure to programming languages came through learning Java, which isn’t the most beginner-friendly language. However, after years of avoidance, I embarked on a self-learning journey to master Python. To my surprise, I found it more approachable than Java, and I began to thoroughly enjoy it. Despite this newfound appreciation, I lacked opportunities to apply my skills in real-time. Fast-forward to joining the Data Analytics class and club at LBS, where Machine Learning was demystified. The icing on the cake? Python is the primary programming language applied. Oh, happy days!
Machine learning is a subset of artificial intelligence that focuses on creating systems and algorithms capable of learning and improving from experience without being explicitly programmed. It falls under the broader category of predictive analytics within the field of data analytics. Predictive analytics involves using various statistical algorithms and machine learning techniques to analyze historical data, identify patterns, and make predictions about future events or outcomes. The key components of machine learning include data, algorithms, and model training (data modeling), leading to systems that can generalize patterns and make informed decisions.
In the context of data analytics, machine learning can be categorized into Supervised and Unsupervised Machine Learning.
Supervised Machine Learning:
Supervised machine learning learns patterns and relationships between input and output data. It utilizes labeled data which are datasets that contain a lot of examples of Features and Target. It gathers information based on a model, and applies algorithm to produce pattern, which are then used for prediction and eventually deployed to take decision. This model can now be applied to new and similar cases.
There are two types of supervised learning algorithms:
Classification: Categorizes data into predefined classes or groups based on qualitative features. For example:
A bank may have a customer dataset containing credit history, loans, investment details, etc. and they may want to know if any customer will default. In the historical data, we will have Features and Target.
Features will be attributes of a customer, such as credit history, loans, investments, etc.
Target will represent whether a particular customer has defaulted in the past (normally represented by 1 or 0 / True or False / Yes or No.
Classification algorithms are used for predicting distinct outcomes, if the outcome can take only two possible values such as True or False, it is known as Binary Classification. When the outcome contains more than two possible values and does not involve comparison, it is known as Multiclass Classification, e.g.religion. When the variable is measured on an arbitrary scale, it is known as Ordinal Classification, e.g. Agree/Neutral/Disagree.
Regression Analysis: Predicts a continuous variable based on historical data, allowing for the identification of trends and patterns. For example:
In a dataset containing features of a house such as lot size, number of bedrooms, number of baths, neighborhood, etc. and the price of the house, a Regression algorithm can be trained to learn the relationship between the features and the price of the house.
Regression deals with pattern recognition and algorithms used here are Linear regression, Decision Tree Regression, etc. The datasets used are typically quantitative in nature.
By leveraging machine learning algorithms, data analysts can extract valuable insights, automate decision-making processes, and enhance the overall predictive capabilities of their analytics solutions.
My First Job As An Entry Level Investment Banker