527,856 Hits in 3.7 sec

Precision and Recall for Time Series [article]

Nesime Tatbul, Tae Jun Lee, Stan Zdonik, Mejbah Alam, Justin Gottschlich
2019 arXiv   pre-print
Our model expands the well-known Precision and Recall metrics to measure ranges, while simultaneously enabling customization support for domain-specific preferences.  ...  Motivated by this observation, we present a new mathematical model to evaluate the accuracy of time series classification algorithms.  ...  Acknowledgments We thank Eric Metcalf for his help with experiments. This research has been funded in part by Intel and by NSF grant IIS-1526639.  ... 
arXiv:1803.03639v3 fatcat:w7ullujqcvamzagalxh7va6r7i

Model-based search in large time series databases

Alexios Kotsifakos, Vassilis Athitsos, Panagiotis Papapetrou, Jaakko Hollmén, Dimitrios Gunopulos
2011 Proceedings of the 4th International Conference on PErvasive Technologies Related to Assistive Environments - PETRA '11  
such sensors over large time intervals.  ...  In particular, we describe the two different approaches, and we identify some important pros and cons for each approach.  ...  to Professors Chris Ding and Fillia Makedon.  ... 
doi:10.1145/2141622.2141666 dblp:conf/petra/KotsifakosAPHG11 fatcat:br7yeelpsrfkzpztotkntjndf4

Estimating the Best Time to View Cherry Blossoms Using Time-Series Forecasting Method

Tomonari Horikawa, Munenori Takahashi, Masaki Endo, Shigeyoshi Ohno, Masaharu Hirota, Hiroshi Ishikawa
2022 Machine Learning and Knowledge Extraction  
This study proposes a time-series prediction method using SNS data and machine learning as a new method for estimating the best times for viewing for a certain period.  ...  Combining the time-series forecasting method and the low-cost moving average method yields an estimate of the best time to view cherry blossoms.  ...  data, and by the values of recall and precision.  ... 
doi:10.3390/make4020018 fatcat:7un42nag7vhnfd2jcom6mglw6i

Benchmarking Deep Learning Interpretability in Time Series Predictions [article]

Aya Abdelsalam Ismail, Mohamed Gunady, Héctor Corrada Bravo, Soheil Feizi
2020 arXiv   pre-print
We propose and report multiple metrics to empirically evaluate the performance of saliency methods for detecting feature importance over time using both precision (i.e., whether identified features contain  ...  These methods are mostly used in vision and language tasks, and their applications to time series data is relatively unexplored.  ...  Acknowledgements We thank Kalinda Vathupola for his thoughtful feedback on this work.  ... 
arXiv:2010.13924v1 fatcat:4c7hmjrgfjchdb6b4py22yspey

Automatic SARIMA Order Identification Convolutional Neural Network

Paisit Khanarsa, Department of Mathematics and Computer Science, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
2020 International Journal of Machine Learning and Computing  
For the time series models such as the autoregressive integrated moving average (ARIMA) model and the seasonal autoregressive integrated moving average (SARIMA) model, statisticians mostly identify the  ...  The performance of the ASOC model provides better performance than the likelihood method via precision, recall and f1-score.  ...  Fig. 7 demonstrates precision, recall and f1-score between model A1 and model C which it is clear that model A1 gives the best precision, recall and f1-score. C.  ... 
doi:10.18178/ijmlc.2020.10.5.988 fatcat:tuya5tuslbfspfxv4zsb4sdn6u

Precision and Recall for Range-Based Anomaly Detection [article]

Tae Jun Lee, Justin Gottschlich, Nesime Tatbul, Eric Metcalf, Stan Zdonik
2018 arXiv   pre-print
In this paper, we present a new mathematical model to express range-based anomalies, anomalies that occur over a range (or period) of time.  ...  Precision and Recall for Range-Based Anomaly Detection SysML'18, February 2018, Stanford, CA, USA  ...  correctness for domain-specific time-series anomalies.  ... 
arXiv:1801.03175v3 fatcat:ragils3mqfa5depcm4aizzh3eq

Using Relevance Feedback to Learn Both the Distance Measure and the Query in Multimedia Databases [chapter]

Chotirat Ann Ratanamahatana, Eamonn Keogh
2005 Lecture Notes in Computer Science  
system and query refinement can further improve the precision/recall by a wide margin.  ...  We demonstrate utility of our approach on both classification and query retrieval tasks for time series and other types of multimedia data, then show that its incorporating into the relevance feedback  ...  the trade off between efficiency and precision/recall.  ... 
doi:10.1007/11552451_3 fatcat:7be63scew5ca3albix4b3nuwua

Unsupervised Anomaly Detection Approach for Time-Series in Multi-Domains Using Deep Reconstruction Error

Tsatsral Amarbayasgalan, Van Huy Pham, Nipon Theera-Umpon, Keun Ho Ryu
2020 Symmetry  
We conducted two types of experiments on a total of 52 publicly available time-series benchmark datasets for the batch and real-time anomaly detections.  ...  Automatic anomaly detection for time-series is critical in a variety of real-world domains such as fraud detection, fault diagnosis, and patient monitoring.  ...  We compared the precision, recall, and F-measure of the RE-ADTS to the evaluation of six algorithms on 20 time-series datasets in [20] .  ... 
doi:10.3390/sym12081251 fatcat:bm5cjz7775f5tnmjjfphvdldca

Learning stochastic finite-state transducer to predict individual patient outcomes

Patricia Ordoñez, Nelson Schwarz, Adnel Figueroa-Jiménez, Leonardo A. Garcia-Lebron, Abiel Roche-Lima
2016 Health and Technology  
The high frequency data in intensive care unit is flashed on a screen for a few seconds and never used again.  ...  In this paper, we learned the edit distance costs of a symbolic univariate time series representation through a stochastic finite-state transducer to predict patient outcomes in intensive care units.  ...  reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.  ... 
doi:10.1007/s12553-016-0146-2 pmid:27942425 pmcid:PMC5124435 fatcat:6yqc64yo4vejxb4tvxfxen2oyu

An approach for dynamical network reconstruction of simple network motifs

Masahiko Nakatsui, Michihiro Araki, Akihiko Kondo
2013 BMC Systems Biology  
As a result, our method could reconstruct about 40% of interactions in network motif from time-series data set.  ...  Moreover the introduction of time-series data of one-factor disrupted model could remarkably improved the performance of network inference.  ...  Acknowledgements This work was partly supported by the commission for Development of Artificial Gene Synthesis Technology for Creating Innovative Biomaterial from the Ministry of Economy, Trade and Industry  ... 
doi:10.1186/1752-0509-7-s6-s4 pmid:24564905 pmcid:PMC4029519 fatcat:l6dicttqyre57gl4r5ya2ynlwu

Granger Causality for Time-Series Anomaly Detection

Huida Qiu, Yan Liu, Niranjan A. Subrahmanya, Weichang Li
2012 2012 IEEE 12th International Conference on Data Mining  
Recent developments in industrial systems provide us with a large amount of time series data from sensors, logs, system settings and physical measurements, etc.  ...  However, the special characteristics of these time series data, such as high dimensions and complex dependencies between variables, as well as its massive volume, pose great challenges to existing anomaly  ...  1 = 2 · Recall · Precision/(Recall + Precision)) of all eight methods.  ... 
doi:10.1109/icdm.2012.73 dblp:conf/icdm/QiuLSL12 fatcat:qf2xrcisezdwvgxrjuaftwnucm

All-Clear Flare Prediction Using Interval-based Time Series Classifiers [article]

Anli Ji, Berkay Aydin, Manolis K. Georgoulis, Rafal Angryk
2021 arXiv   pre-print
Our results show that time series classifiers provide better forecasting results in terms of skill scores, precision and recall metrics, and they can be further improved for more precise all-clear forecasts  ...  Our study focuses on training and testing a set of interval-based time series classifiers named Time Series Forest (TSF).  ...  Results are taken from undersampled datasets (Partition 1 and 2 for Base Learners and Partition 4 for Meta-learner) Precision(XM) TPR/Recall(XM) Precision(CBN) TNR/Recall(CBN) The precision and recall  ... 
arXiv:2105.01202v1 fatcat:yjrtzspqprbhvnbqt6zqldably

Comparison of Recurrent Neural Network Architectures for Wildfire Spread Modelling [article]

Rylan Perumal, Terence L van Zyl
2020 arXiv   pre-print
Overall the GRU performs better for longer time series than the LSTM.  ...  Through active fire analysis, it is possible to reproduce a dynamical process, such as wildfires, with limited duration time series data.  ...  ACKNOWLEDGMENT I would like to thank Prof. van Zyl for all of his intellectual insight and motivation throughout. It was a great pleasure being supervised by him.  ... 
arXiv:2005.13040v1 fatcat:mos3pybpmvczplgqblcyrzahra


Milica Ćirić, Bratislav Predić
2021 Facta Universitatis Series Automatic Control and Robotics  
This process was then repeated for the entire six month period and a slight downward trend can be noticed for error metrics, leading to the conclusion that the network would perform even better over time  ...  By aggregating purchase data for all products a customer purchased, we were able to get more precise predictions of the next purchase.  ...  For these two classes precision and recall were calculated and then compared to earlier results.  ... 
doi:10.22190/fuacr2003151c fatcat:rc4xelcqmrdnhh7zfgel3rmqqi

Statistical Evaluation of Anomaly Detectors for Sequences [article]

Erik Scharwächter, Emmanuel Müller
2020 arXiv   pre-print
In this work, we formalize a notion of precision and recall with temporal tolerance for point-based anomaly detection in sequential data.  ...  Although precision and recall are standard performance measures for anomaly detection, their statistical properties in sequential detection settings are poorly understood.  ...  They introduced novel precision and recall measures for range-based anomaly detection.  ... 
arXiv:2008.05788v1 fatcat:r3zbmawqkzhzxpsxxctpfet5hm
« Previous Showing results 1 — 15 out of 527,856 results