2022
DOI: 10.1007/s12650-022-00890-3
|View full text |Cite
|
Sign up to set email alerts
|

Representation and analysis of time-series data via deep embedding and visual exploration

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

1
3
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
3
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(4 citation statements)
references
References 39 publications
1
3
0
Order By: Relevance
“…According to the respective results and focusing on largest pattern sizes (i.e., sets of co‐occurring categories), we can describe the following profile of a VA technique involving embeddings: designed for the AI/ML domain, using neural network approaches for embedding computations (while not specifying explicitly the size of respective embedding vectors), supporting the visual analytic task of comparison/selection, explicit visual representation of embeddings (or derived results) and interaction with them, while discussing evaluation of such human‐centered aspects in the respective publication. This pattern is supported by 9 entries (7% of the complete data set) in our current survey data [FZCM20, GHM21, SKB*21, CZG*22,PdSP*22,RSL*22,ZJQH22,HHS*23,WHC*23]. Shorter patterns occur more frequently, up to 15 entries (12%) supporting the pattern of the AI/ML domain, text data, neural networks, and interactive visualization & evaluation concerns .…”
Section: Survey Data Analysessupporting
confidence: 69%
See 2 more Smart Citations
“…According to the respective results and focusing on largest pattern sizes (i.e., sets of co‐occurring categories), we can describe the following profile of a VA technique involving embeddings: designed for the AI/ML domain, using neural network approaches for embedding computations (while not specifying explicitly the size of respective embedding vectors), supporting the visual analytic task of comparison/selection, explicit visual representation of embeddings (or derived results) and interaction with them, while discussing evaluation of such human‐centered aspects in the respective publication. This pattern is supported by 9 entries (7% of the complete data set) in our current survey data [FZCM20, GHM21, SKB*21, CZG*22,PdSP*22,RSL*22,ZJQH22,HHS*23,WHC*23]. Shorter patterns occur more frequently, up to 15 entries (12%) supporting the pattern of the AI/ML domain, text data, neural networks, and interactive visualization & evaluation concerns .…”
Section: Survey Data Analysessupporting
confidence: 69%
“…We classify the computational approach into three main methods, which can of course be used in combination within the scope of a single paper. The most common, with 88 occurrences, is neu ral network, where different types of ML/DL models are used to obtain the embeddings [ZJQH22,RSL*22, SDXR22]. Even though there are many different variants, some of which are highly customized for the specific tasks, our analysis shows that using the BERT [DCLT19] model or word2vec [MSC*13] (or some of the other versions inspired/derived from it) is a common choice among the surveyed papers.…”
Section: Categorization Of Va + Embedding Approachesmentioning
confidence: 99%
See 1 more Smart Citation
“…The representation of time-series data is crucial for several analytical tasks, including comparison analysis, clustering, and classification. Conventional illustration methods yield uniform representations derived from statistical data [1]. Moreover, analyzing time-series data on currency values such as bitcoin can provide insights for investment decisions.…”
Section: Introductionmentioning
confidence: 99%