Analyze Shapley Values with Altair® Knowledge Studio®

Knowledge Studio’s Weight of Evidence node is essential to creating scorecard models. The Weight of Evidence node transforms your data so the model can assign points to every bin and these points will eventually add up to the scorecard. Knowledge Studio uses Altair’s patented decision trees to bin ordinal and some continuous variables. If the bins don’t make business sense, you can optimize them automatically by binning them for monotonicity and disallowing pure nodes. Knowledge Studio can optimize all variables in your dataset with a single click. You can also edit the splits if you have business reasons to bin your variables differently. Once Knowledge Studio has transformed your variables, you can see the code that created the transformation. Click here to learn more about Altair Knowledge Studio.

This series of short videos will help you understand the basics of data science. In Part 1, we focus on predictive analytics, which is the process of training computers with historical data so they can make accurate predictions. Click here to learn more about machine learning.

This is part two in a series of short videos to help you understand the basics of data science. This video focuses on the concepts of supervised and unsupervised learning. In supervised learning, you develop algorithms that predict outcomes based on independent variables and you use historical data to train your algorithms. Unsupervised learning models identify hidden patterns in your data and group records into clusters; they act on the dataset without human guidance. Click here to learn more about machine learning.

This is part three in a series of short videos to help you understand the basics of data science. This video focuses on prescriptive analytics, which uses results from multiple machine learning algorithms to inform future decisions. With prescriptive analytics, the object is to optimize and, to some extent, automate the decision-making process. A prescriptive analytics workflow can use multiple algorithms to prescribe actions based on the characteristics of new data. Click here to learn more about machine learning.

Knowledge Studio’s novelty and outlier detector node makes it easy to identify anomalies in a dataset and remove them if desired. The software supports three different methods for detecting outliers: Isolation Forest, Local Outlier Factor, and One Class SVM. This video demonstrates how to use the novelty and outlier detector. Click here to learn more about Altair Knowledge Studio.

XGB stands for “eXtreme Gradient Boosting” and is often referred to as XGBoost. Knowledge Studio includes an XGB node for predictive modeling that data scientists can use to develop solutions for classification and regression problems. Knowledge Studio’s XGB implementation supports models with several types of dependent variables and can handle single numeric DV, single binary DV, single multiclass DV, multiple numeric DVs, and multiple binary DVs. This video shows how to use Knowledge Studio’s XGB node. Click here to learn more about Altair Knowledge Studio.

Knowledge Studio supports Autoregressive Integrated Moving Average (ARIMA) models, a powerful way to make accurate predictions based on time series data. You can add ARIMA models to your AI workflows with a fully menu-driven user interface. The software’s Auto ARIMA functions automatically estimate values for ARIMA parameters using a grid search or step-wise algorithm. ARIMA is a simple yet powerful method for making time series forecasts, often incorporating seasonal and other types of semi-regular variations. For example, you can use ARIMA models to forecast electricity and raw materials utilization in a factory, output volumes in an oil refinery, fuel consumption for truck fleet, rail, and seaborne shipping companies, patient churn and intake volumes in hospitals, and key financial indicators for any type of business. Click here to learn more about Altair Knowledge Studio.

Datasets often have missing values due to file corruption, failure to record data points, or other causes. Handling missing data values correctly is critical to developing accurate predictive models. Knowledge Studio makes it easy to identify datasets containing missing values and generate new substitute values based on a variety of substitution algorithms. This video walks you through a simple example of how the software’s Substitute Missing Values node works. Click here to learn more about Altair Knowledge Studio.

In simple terms, Simpson’s Paradox occurs when a trend appears in subgroups but disappears or is reversed when subgroups are combined into a single dataset. Knowledge Studio supports detection of this statistical phenomenon. In this video, you will see an example of how Simpson’s Paradox can manifest itself and how you can use Knowledge Studio to detect its presence automatically. Click here to learn more about Altair Knowledge Studio.

Most machine learning algorithms assume there are equal numbers of examples for each class in the source data. Many datasets contain substantially different numbers of records for important classes — resulting in an imbalanced class problem. Failure to handle this properly results in models with poor predictive performance. Knowledge Studio has a node specifically built to handle imbalanced class issues. In this video, you will learn how to identify an imbalanced class problem and use the software’s Handle Class Imbalance node to correct it. Refer to the Imbalanced-Learn Documentation website to learn more about the challenges related to working with imbalanced classes

In the context of machine learning applications, GLM models allows the use of dependent variables that do not follow normal distributions. This video shows how easy it is to use Knowledge Studio’s GLM node to utilize this advanced statistical technique to build more accurate machine learning models. Click here to learn more about Altair Knowledge Studio.