Discovery of motifs to forecast outlier occurrence in time series.
Francisco Martínez-ÁlvarezAlicia Troncoso LoraJosé C. RiquelmeJesús S. Aguilar-RuizPublished in: Pattern Recognit. Lett. (2011)
Keyphrases
- motif discovery
- motif discovery algorithm
- recurring patterns
- dna sequences
- motif finding
- pattern discovery
- biological sequences
- data mining tasks
- computational biology
- binding sites
- multiobjective optimization
- forecasting accuracy
- box jenkins
- exponential smoothing
- mdl principle
- temporal patterns
- short term
- chaotic time series
- stock market
- outlier detection
- multivariate time series
- phase space reconstruction
- surprising patterns
- arima model
- biological networks
- novelty detection
- artificial neural networks
- stock price
- hybrid model
- dynamic time warping
- association rule mining
- non stationary
- long term