Unsupervised mining of long time series based on latent topic model

This paper presents a novel unsupervised method for mining time series based on two generative topic models, i.e., probabilistic Latent Semantic Analysis (pLSA) and Latent Dirichlet Allocation (LDA). The proposed method treats each time series as a text document, and extracts a set of local patterns...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) Vol. 103; pp. 93 - 103
Main Authors: Wang, Jin, Sun, Xiangping, She, Mary F.H., Kouzani, Abbas, Nahavandi, Saeid
Format: Journal Article
Language:English
Published: Amsterdam Elsevier B.V 01.03.2013
Elsevier
Subjects:
Applied sciences
Artificial intelligence
Bag-of-patterns
Biological and medical sciences
Categories
Computer science; control theory; systems
Data processing. List processing. Character string processing
Dirichlet problem
ECG signals
Electrocardiography. Vectocardiography
Electrodiagnosis. Electric activity recording
Exact sciences and technology
Inference from stochastic processes; time series analysis
Investigative techniques, diagnostic techniques (general aspects)
LDA
Mathematical models
Mathematics
Medical sciences
Memory organisation. Data processing
Mining
pLSA
Probability and statistics
Sciences and techniques of general use
Semantics
Software
Speech and sound recognition and synthesis. Linguistics
Statistics
Texts
Time series
Unsupervised learning
LDA
ECG signals
Bag-of-patterns
Unsupervised learning
pLSA
Text analysis
K means algorithm
Algebraic semantic
Data mining
Modeling
Document processing
Business process
Graph decomposition
Bag of words
Classification
Sliding window
Database
Electrocardiography
Linguistics
Natural language
Categorization
Document analysis
Generative model
Concept learning
Probabilistic approach
Time series
Text
Experimental result
Document structure
ISSN:0925-2312, 1872-8286
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a novel unsupervised method for mining time series based on two generative topic models, i.e., probabilistic Latent Semantic Analysis (pLSA) and Latent Dirichlet Allocation (LDA). The proposed method treats each time series as a text document, and extracts a set of local patterns from the sequence as words by sliding a short temporal window along the sequence. Motivated by the success of latent topic models in text document analysis, latent topic models are extended to find the underlying structure of time series in an unsupervised manner. The clusters or categories of unlabeled time series are automatically discovered by the latent topic models using bag-of-patterns representation. The proposed method was experimentally validated using two sets of time series data extracted from a public Electrocardiography (ECG) database through comparison with the baseline k-means and the Normalized Cuts approaches. In addition, the impact of the bag-of-patterns' parameters was investigated. Experimental results demonstrate that the proposed unsupervised method not only outperforms the baseline k-means and the Normalized Cuts in learning semantic categories of the unlabeled time series, but also is relatively stable with respect to the bag-of-patterns' parameters. To the best of our knowledge, this work is the first attempt to explore latent topic models for unsupervised mining of time series data. ► A novel unsupervised method based on latent topic modelis proposedfor mining time series. ► The proposed method processes time series as text documents. ► The topic model based method is able to effectively capture structural similarity information. ► The proposed method can be potentially used for time series segmentation.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2012.09.008