ADS: the adaptive data series index

Zoumpatianos, Kostas; Idreos, Stratos; Palpanas, Themis

doi:10.1007/s00778-016-0442-5

Cited by 72 publications

(104 citation statements)

References 40 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…Observe that when δ = 1, a δ--approximate method becomes -approximate, and when = 0, an -approximate method becomes exact [21]. It it also possible that the same approach implements both approximate and exact algorithms [73,81,14,89,71]. Methods that provide exact answers with probabilistic guarantees are considered δ-0-approximate.…”

Section: Definitions and Terminologymentioning

confidence: 99%

“…Nevertheless, despite the significance of data series similarity search, and the abundance of relevant methods that have been proposed in the past two decades [3,22,73,63,23,13,42,66,71,81,14,89,58,85,52,51,62], no study has ever attempted to compare these methods under the same conditions. We also point out that we focus on the efficiency of similarity search methods, whereas previous works studied the accuracy of dimensionality reduction techniques and similarity measures, focusing on classification [44,27,9].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The lernaean hydra of data series similarity search

et al. 2018

Self Cite

View full text Add to dashboard Cite

Increasingly large data series collections are becoming commonplace across many different domains and applications.A key operation in the analysis of data series collections is similarity search, which has attracted lots of attention and effort over the past two decades. Even though several relevant approaches have been proposed in the literature, none of the existing studies provides a detailed evaluation against the available alternatives. The lack of comparative results is further exacerbated by the non-standard use of terminology, which has led to confusion and misconceptions. In this paper, we provide definitions for the different flavors of similarity search that have been studied in the past, and present the first systematic experimental evaluation of the efficiency of data series similarity search techniques. Based on the experimental results, we describe the strengths and weaknesses of each approach and give recommendations for the best approach to use under typical use cases. Finally, by identifying the shortcomings of each method, our findings lay the ground for solid further developments in the field. PVLDB Reference Format:the whole (not a sub-) sequence. This problem represents a common use case across many domains [1, 2,38,29]. This work is the most extensive experimental comparison of the efficiency of similarity search methods ever conducted. Contributions. We make the following contributions:1. We present a thorough discussion of the data series similarity search problem, formally defining its different variations that have been studied in the literature under diverse and conflicting names. Thus, establishing a common language that will facilitate further work in this area.2. We include a brief survey of data series similarity search approaches, bringing together studies presented in different communities that have been treated in isolation from each other. These approaches range from smart serial scan methods to the use of indexing, and are based on a variety of classic and specialized data summarization techniques.3. We make sure that all approaches are evaluated under the same conditions, so as to guard against implementation bias. To this effect, we used implementations in C/C++ for all approaches, and reimplemented in C the ones that were only available in other programming languages. Moreover, we conducted a careful inspection of the code bases, and applied to all of them the same set of optimizations (e.g., with respect to memory management, Euclidean distance calculation, etc.), leading to considerably faster performance.4. We conduct the first comprehensive experimental evaluation of the efficiency of data series similarity search approaches, using several synthetic and 4 real datasets from diverse domains. In addition, we report the first large scale experiments with carefully crafted query workloads that include queries of varying difficulty, which can effectively stress-test all the approaches. Our results reveal characteristics that have not been reported in the literature, and lead...

show abstract

Section: Definitions and Terminologymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The lernaean hydra of data series similarity search

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Time series 1 have gathered the attention of the data management community for more than two decades [1,14,27]. They are one of the most common data types, present in virtually every scientific and social domain: they appear as audio sequences [12], shape and image data [26], financial [23], environmental monitoring [22] and scientific data [10], and they have many diverse applications, such as in health care, astronomy, biology, economics, etc.…”

Section: Introductionmentioning

confidence: 99%

Report on the First and Second Interdisciplinary Time Series Analysis Workshop (ITISA)

Palpanas

Beckmann

2019

SIGMOD Rec.

Self Cite

View full text Add to dashboard Cite

The analysis of time-series data associated with modernday industrial operations and scientific experiments is now pushing both computational power and resources to their limits. In order to analyze the existing and (more importantly) future very large time series collections, new technologies and the development of more efficient and smarter algorithms are required. The two editions of the Interdisciplinary Time Series Analysis Workshop brought together data analysts from the fields of computer science, astrophysics, neuroscience, engineering, electricity networks, and music. The focus of these workshops was on the requirements of different applications in the various domains, and also on the advances in both academia and industry, in the areas of time-series management and analysis. In this paper, we summarize the experiences presented in and the results obtained from the two workshops, highlighting the relevant state-ofthe-art-techniques and open research problems.

show abstract

“…Note that previous studies have shown that the iSAX index is robust with respect to the choice of parameters (word length, cardinality, leaf threshold) [16], [5], [21]. Moreover, it can also be used to answer queries with the Dynamic Time Warping (DTW) distance, through the use of the corresponding lower bounding envelope [9].…”

Section: A Isax Representationmentioning

confidence: 99%

“…Although recent studies have shown that in certain cases sequential scans can be very efficient [14], [18], such techniques are only advantageous when the database consists of a single, long time series, and query answers are small subsequences of this long time series. Such approaches, however, are not beneficial in the general case of querying a mixed database of many small time series [21] (e.g., in neuroscience, or manufacturing applications [12]), which is the focus of this * The research leading to these results has received funding from the Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 (a) Straightforward implementation: the batch of queries is duplicated on all the computing nodes. study.…”

Section: Introductionmentioning

confidence: 99%

DPiSAX: Massively Distributed Partitioned iSAX

Yagoubi

Akbarinia

Masséglia

et al. 2017

2017 IEEE International Conference on Data Mining (ICDM)

Self Cite

View full text Add to dashboard Cite

Abstract-Indexing is crucial for many data mining tasks that rely on efficient and effective similarity query processing. Consequently, indexing large volumes of time series, along with high performance similarity query processing, have became topics of high interest. For many applications across diverse domains though, the amount of data to be processed might be intractable for a single machine, making existing centralized indexing solutions inefficient. We propose a parallel indexing solution that gracefully scales to billions of time series, and a parallel query processing strategy that, given a batch of queries, efficiently exploits the index. Our experiments, on both synthetic and real world data, illustrate that our index creation algorithm works on 1 billion time series in less than 2 hours , while the state of the art centralized algorithms need more than 5 days. Also, our distributed querying algorithm is able to efficiently process millions of queries over collections of billions of time series, thanks to an effective load balancing mechanism.

show abstract

ADS: the adaptive data series index

Cited by 72 publications

References 40 publications

The lernaean hydra of data series similarity search

The lernaean hydra of data series similarity search

Report on the First and Second Interdisciplinary Time Series Analysis Workshop (ITISA)

DPiSAX: Massively Distributed Partitioned iSAX

Contact Info

Product

Resources

About