Keep it Simple : Machine Learning & Algorithms for Big Boys ( too old to Learn they say...)

Let's Start from the very beginning . What the Hell is Machine Learning ? (Thats how I asked the question during one of my interaction with a Data Scientist few months back

Machine learning is a data science technique that allows computers to use existing data to forecast future behaviors, outcomes, and trends. Using machine learning, computers learn without being explicitly programmed.Forecasts or predictions from machine learning can make apps and devices smarter.

• When you shop online, machine learning helps recommend other products you might like based on what you've purchased.
• When your credit card is swiped, machine learning compares the transaction to a database of transactions and helps detect fraud.
• When your robot vacuum cleaner vacuums a room, machine learning helps it decide whether the job is done.

The second question to the gentleman is

"What are the Flavors of machine learning ? " and he politely narrated

1.Supervised Learning

Supervised learning algorithms make predictions based on a set of examples. For instance, historical stock prices can be used to hazard guesses at future prices. Each example used for training is labeled with the value of interest—in this case the stock price. A supervised learning algorithm looks for patterns in those value labels. It can use any information that might be relevant—the day of the week, the season, the company's financial data, the type of industry, the presence of disruptive geopolitical events—and each algorithm looks for different types of patterns. After the algorithm has found the best pattern it can, it uses that pattern to make predictions for unlabeled testing data—tomorrow's prices.

Supervised learning is a popular and useful type of machine learning. There are several specific types of supervised learning like classification, regression, and anomaly detection.

• Classification. When the data are being used to predict a category, supervised learning is also called classification. This is the case when assigning an image as a picture of either a 'cat' or a 'dog'. When there are only two choices, it's called two-class or binomial classification. When there are more categories, as when predicting the winner of the Golf tournament, this problem is known as multi-class classification.
• Regression. When a value is being predicted, as with stock prices, supervised learning is called regression.
• Anomaly detection. Sometimes the goal is to identify data points that are simply unusual. In fraud detection, for example, any highly unusual credit card spending patterns are suspect. The possible variations are so numerous and the training examples so few, that it's not feasible to learn what fraudulent activity looks like. The approach that anomaly detection takes is to simply learn what normal activity looks like (using a history non-fraudulent transactions) and identify anything that is significantly different.

2.Unsupervised Learning

In unsupervised learning, data points have no labels associated with them. Instead, the goal of an unsupervised learning algorithm is to organize the data in some way or to describe its structure. This can mean grouping it into clusters or finding different ways of looking at complex data so that it appears simpler or more organized.

3.Reinforcement learning

In reinforcement learning, the algorithm gets to choose an action in response to each data point. The learning algorithm also receives a reward signal a short time later, indicating how good the decision was. Based on this, the algorithm modifies its strategy in order to achieve the highest reward. Reinforcement learning is common in robotics, where the set of sensor readings at one point in time is a data point, and the algorithm must choose the robot's next action. It is also a natural fit for Internet of Things applications.

I asked him about what are the other common machine learning terms ' just to make sure I use those Jargons :)

• Algorithm: A self-contained set of rules used to solve problems through data processing, math, or automated reasoning.
• Anomaly detection: A model that flags unusual events or values and helps you discover problems. For example, credit card fraud detection looks for unusual purchases.
• Categorical data: Data that is organized by categories and that can be divided into groups. For example a categorical data set for autos could specify year, make, model, and price.
• Classification: A model for organizing data points into categories based on a data set for which category groupings are already known.
• Feature engineering: The process of extracting or selecting features related to a data set in order to enhance the data set and improve outcomes. For instance, flight data could be enhanced by days of the week and holidays.
• Module: A functional part in a Machine Learning Studio model, such as the Enter Data module that enables entering and editing small data sets. An algorithm is also a type of module in any Machine Learning Studio.
• Model: A supervised learning model is the product of a machine learning experiment comprised of training data, an algorithm module, and functional modules, such as a Score Model module.
• Numerical data: Data that has meaning as measurements (continuous data) or counts (discrete data). Also referred to as quantitative data.
• Partition: The method by which you divide data into samples.
• Prediction: A prediction is a forecast of a value or values from a machine learning model. You might also see the term "predicted score." However, predicted scores are not the final output of a model. An evaluation of the model follows the score.
• Regression: A model for predicting a value based on independent variables, such as predicting the price of a car based on its year and make.
• Score: A predicted value generated from a trained classification or regression model, using the Score model in Machine Learning Studio. Classification models also return a score for the probability of the predicted value. Once you've generated scores from a model, you can evaluate the model's accuracy using the Evaluate the model module
• Sample: A part of a data set intended to be representative of the whole. Samples can be selected randomly or based on specific features of the data set.

All these explanations are good but tell me How to choose algorithms for M Machine Learning

He said ,The answer to the question "What machine learning algorithm should I use?" is always "It depends." It depends on the size, quality, and nature of the data. It depends on what you want to do with the answer. It depends on how the math of the algorithm was translated into instructions for the computer you are using. And it depends on how much time you have. Even the most experienced data scientists can't tell which algorithm will perform best before trying them.

He showed me an example with Microsoft Azure Machine Learning

The Microsoft Azure Machine Learning Algorithm Cheat Sheet helps you choose the right machine learning algorithm for your predictive analytics solutions from the Microsoft Azure Machine Learning library of algorithms. The Picture below is a quick snapshot view of the same

Then my question to him is " How to use the cheat sheet" He said ..

Read the path and algorithm labels on the chart as "For <path label>, use <algorithm>." For example, "For speed, use two class logistic regression." Sometimes more than one branch applies. Sometimes none of them are a perfect fit. They're intended to be rule-of-thumb recommendations, so don't worry about it being exact. Several data scientists I talked with said that the only sure way to find the very best algorithm is to try all of them. Ooops I said :)

What are the other Considerations when choosing an algorithm, I asked.

Accuracy

Getting the most accurate answer possible isn't always necessary. Sometimes an approximation is adequate, depending on what you want to use it for. If that's the case, you may be able to cut your processing time dramatically by sticking with more approximate methods. Another advantage of more approximate methods is that they naturally tend to avoid overfitting.

Training time

The number of minutes or hours necessary to train a model varies a great deal between algorithms. Training time is often closely tied to accuracy—one typically accompanies the other. In addition, some algorithms are more sensitive to the number of data points than others. When time is limited it can drive the choice of algorithm, especially when the data set is large.

Linearity

Lots of machine learning algorithms make use of linearity. Linear classification algorithms assume that classes can be separated by a straight line (or its higher-dimensional analog). These include logistic regression and support vector machines (as implemented in Azure Machine Learning). Linear regression algorithms assume that data trends follow a straight line. These assumptions aren't bad for some problems, but on others they bring accuracy down.

Non-linear class boundary - relying on a linear classification algorithm would result in low accuracy.Data with a nonlinear trend - using a linear regression method would generate much larger errors than necessary.Despite their dangers, linear algorithms are very popular as a first line of attack. They tend to be algorithmically simple and fast to train.

I said stop !!! I am Too old to Read more :) , Lets cover the rest in the 2nd Part .

Thank you for your Patience :)

Regard

Dinesh Chandrasekar DC