Page 278 - AI Computer 10

Unsupervised widgets help identify patterns and structures in unlabeled data.

Widget Category: Unsupervised Widget Name Description
Visualises the distances or dissimilarities between
Distance Matrix
data points in a dataset.

Correlations Computes all pairwise attribute correlations.

k-Means Groups items using the k-Means clustering algorithm.
Visualises distances between objects using coloured
Distance Map
spots.
Stage: Evaluation
Evaluation widgets are used to assess the performance and effectiveness of machine learning models. These
widgets allow users to validate predictions, compare models, and visualize results based on various evaluation
metrics.

Widget Category: Evaluation Widget Name Description
Test and Score Evaluates the performance of ML models by applying
them to a test dataset.
Predictions Visualises predictions based on trained ML models,
making it an essential part of the workflow for
predictive analysis.
Confusion Matrix Visualises the model's performance by comparing
the predicted classifications with the actual
classifications.
ROC Analysis Distinguishes between positive and negative classes
across various threshold settings.
Thus, we can see that the Orange Data Mining tool provides application of No Code AI to all the stages of the AI
Project Cycle.

AI Activity
AI Activity
Statistical Analysis
Palmer Penguins are a species of penguin native to the Antarctic Peninsula area. Research on penguins,
including Palmer Penguins, typically examines their behaviour, habitat, population dynamics, and the
impact of climate change on their ecosystems. Let us learn how to make predictions using Orange Data
Mining.
Step 1: Download the dataset from the following link:
https://www.kaggle.com/code/parulpandey/penguin-dataset-the-new-iris/data.
Step 2: Open the Orange Data Mining app and drag the File widget on the Canvas. Double click the
File icon to upload the training dataset.

144
144