Dataset Analyze Tools

Correlation Matrix Generator

Calculate Pearson, Spearman, or Kendall correlations between numeric columns in a CSV, JSON, or spreadsheet file, with a labelled coefficient matrix, heatmap, and optional significance values.

sales,visits

12,100

18,120

sales ↔ visits

r = 0.98

matrix + heatmap

Dataset Profiler

Profile a CSV, TSV, JSON, JSONL, or spreadsheet file: column types, row/column counts, null and blank rates, unique and duplicate counts, numeric and text statistics, date ranges, top values, and data-quality warnings, with per-column drill-down and CSV/JSON/HTML export.

customers.csv

1,024 rows

columns: 8

nulls, types, duplicates

quality warnings

Fuzzy Duplicate Finder

Find likely duplicate records using configurable text similarity across selected columns. Groups candidates for review and lets you choose which record to keep or merge - nothing is deleted automatically.

Alice Smith, [email protected]

Alicia Smith, [email protected]

likely duplicate

similarity 92%

review group

Outlier Detector

Flag unusually high or low records in numeric columns of a CSV, JSON, or spreadsheet file using IQR, z-score, or modified z-score, with configurable thresholds, optional grouping by a categorical column, and CSV/JSON export of the flagged results.

orders.csv

amount: 12, 15, 999

flagged row 3

amount = 999

severity: high

Time-Series Anomaly Detector

Detect unusual observations in ordered time-series data from a CSV or JSON file using rolling z-score, rolling IQR, or seasonal residual thresholds, with configurable timezone, duplicate, and gap handling. Reports timestamp, series, observed and expected value, score, threshold, and reason - then charts and exports the flagged points without touching your source data.

metrics.csv

2024-01-03, 240

anomaly detected

expected 105

score +4.2