Efficient Retrieval of Similar Time Series

Rafiei, Davood; Mendelzon, Alberto O.

doi:10.1007/978-1-4615-1379-7_6

Cited by 46 publications

(69 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While there were studies in the time series matching and retrieval, they mainly focused on the retrieval from a collection of time series, and not from text data [1,2,3,4,6,8,14,15,16,24,26,27,32]. The main technical challenges for this problem are about the accurate measurement of similarity between two time series, and their efficient search.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Information Retrieval with Time Series Query

Kim

Nikitin

Zhai

et al. 2013

Proceedings of the 2013 Conference on the Theory of Information Retrieval

View full text Add to dashboard Cite

We study a novel information retrieval problem, where the query is a time series for a given time period, and the retrieval task is to find relevant documents in a text collection of the same time period, which contain topics that are correlated with the query time series. This retrieval problem arises in many text mining applications where there is a need to analyze text data in order to discover potentially causal topics. To solve this problem, we propose and study multiple retrieval algorithms that use the general idea of ranking text documents based on how well their terms are correlated with the query time series. Experiment results show that the proposed retrieval algorithm can effectively help users find documents that are relevant to the time series queries, which can help users analyze the variation patterns of the time series.

show abstract

Section: Related Workmentioning

confidence: 99%

“…For more efficient and fast retrieval, various indexing and dimension reduction techniques can also be used, such as discrete Fourier transformation [1,26,27], wavelet transformation [4], and a probabilistic approach [15]. Like the general information retrieval problems, feedback on time series is also studied [16].…”

Section: Related Workmentioning

confidence: 99%

Information Retrieval with Time Series Query

Kim

Nikitin

Zhai

et al. 2013

Proceedings of the 2013 Conference on the Theory of Information Retrieval

View full text Add to dashboard Cite

show abstract

“…The waveform similarity is defined as the similarity of their characteristic value sets, i.e. major FFT coefficients [4], however, each time chunk has an arbitrary initial phase in its oscillation frequency. For instance, two chunks of the same sine wave may have different initial phases, but they should be matched in similarity.…”

Section: Algorithm Modificationsmentioning

confidence: 99%

Similar pattern search for time-sectional oscillation in huge plasma waveform database

Nakanishi

Hochin

Kojima

2006

Fusion Engineering and Design

View full text Add to dashboard Cite

Recently there has been drastic data growth in many fusion plasma experiments. Some computer-aided assistance to find similar waveforms has become indispensable for accelerating data recognition and analysis. A similarity search for slowly varying waveforms, which applies "R-tree" index with Java implementation, was already reported [1]. The next step is the search for time-sectional oscillation patterns. This requires neglecting the initial phase difference between many slices of waveforms. This new algorithm applies power spectrum density (PSD) values instead of FFT complex coefficients. To emphasize the difference between peak PSD frequency of each waveform, its Euclidean distance is multiplied by ωj/ωi (ωj > ωi). By applying "SR-tree" and fast numerical library implemented in C/C++, its computations have been accelerated. This enables the system to deal with much larger data sets. These modifications have successfully extended the application range toward them with verified accuracy.

show abstract

“…In this section we extend the I-kMeans to utilize another equally well-known decomposition method, the Discrete Fourier Transform (DFT) [1,20].…”

Section: I-kmeans With Discrete Fourier Transformmentioning

confidence: 99%

Iterative Incremental Clustering of Time Series

Lin

Vlachos

Keogh

et al. 2004

Lecture Notes in Computer Science

135

View full text Add to dashboard Cite

Abstract. We present a novel anytime version of partitional clustering algorithm, such as k-Means and EM, for time series. The algorithm works by leveraging off the multi-resolution property of wavelets. The dilemma of choosing the initial centers is mitigated by initializing the centers at each approximation level, using the final centers returned by the coarser representations. In addition to casting the clustering algorithms as anytime algorithms, this approach has two other very desirable properties. By working at lower dimensionalities we can efficiently avoid local minima. Therefore, the quality of the clustering is usually better than the batch algorithm. In addition, even if the algorithm is run to completion, our approach is much faster than its batch counterpart. We explain, and empirically demonstrate these surprising and desirable properties with comprehensive experiments on several publicly available real data sets. We further demonstrate that our approach can be generalized to a framework of much broader range of algorithms or data mining problems.

show abstract

Efficient Retrieval of Similar Time Series

Cited by 46 publications

References 10 publications

Information Retrieval with Time Series Query

Information Retrieval with Time Series Query

Similar pattern search for time-sectional oscillation in huge plasma waveform database

Iterative Incremental Clustering of Time Series

Contact Info

Product

Resources

About