Equipment State Classification

Equipment State Classification

Problem Description

You have a dataset containing simulated records from industrial sensors. Each row corresponds to a sample, and the data includes four numerical features:

• Feature1: Vibration measurement on the X-axis.
• Feature2: Vibration measurement on the Y-axis.
• Feature3: Temperature variation measurement.
• Feature4: Pressure variation measurement.

There are four different equipment states (labeled 0, 1, 2, 3). The goal is to assign each sample a label (0, 1, 2, or 3) such that samples from the same state are grouped together. The order of the numbers doesn't matter, only that samples from the same category receive the same label.

Provided Data

• train.csv: Contains training samples with SampleID and the 4 features.
• test.csv: Contains test samples with SampleID and the 4 features. You need to assign labels to these samples.

Objective

Use the training data to understand patterns and relationships between features, then assign each test sample a label between 0 and 3.

Evaluation

Predictions will be evaluated using Adjusted Rand Index (ARI), which measures how well your groupings match the actual (hidden) equipment states.

• ARI = 1.0 indicates perfect match.
• ARI ≈ 0 indicates random match.

Submission File

The file you need to generate after assigning labels to the test set must be in csv format and contain only two columns:

1SampleID,Label21,032,143,05...