hmo_sax: Motif discovery using SAX

In harbinger: A Unified Time Series Event Detection Framework

Motif discovery using SAX

Description

Discovers repeated subsequences (motifs) using a Symbolic Aggregate approXimation (SAX) representation doi:10.1007/s10618-007-0064-z. Subsequences are discretized and grouped by symbolic words; frequently occurring words indicate motifs.

Usage

hmo_sax(a, w, qtd = 2)

Arguments

a	Integer. Alphabet size.
w	Integer. Word/window size.
qtd	Integer. Minimum number of occurrences to classify as a motif.

Value

hmo_sax object.

References

• Lin J, Keogh E, Lonardi S, Chiu B (2007). A symbolic representation of time series, with implications for streaming algorithms. Data Mining and Knowledge Discovery 15, 107–144.

Examples

library(daltoolbox)

# Load motif example data
data(examples_motifs)

# Use a simple sequence example
dataset <- examples_motifs$simple
head(dataset)

# Configure SAX-based motif discovery
model <- hmo_sax(26, 3, 3)

# Fit the model
model <- fit(model, dataset$serie)

# Run detection
detection <- detect(model, dataset$serie)

# Show detected motifs
print(detection[(detection$event),])

harbinger documentation built on Feb. 11, 2026, 9:06 a.m.