GUISSE: A Graphical User Interface For Snippet Selection and Evaluation in Time Series Data

• Guilherme Fernandes Universidade Federal do Ceará (UFC) http://orcid.org/0009-0001-8100-0074
• Lucas Peres Gaspar Universidade Federal do Ceará (UFC)
• Lívia Almada Cruz Universidade Federal do Ceará (UFC)
• Regis Pires Magalhães Universidade Federal do Ceará (UFC)
• José Antônio Macêdo Universidade Federal do Ceará (UFC)

DOI: https://doi.org/10.5753/sbbd_estendido.2025.247671

Resumo

Time series snippet discovery offers powerful tools for summarizing complex temporal data via representative subsequences. While advanced methods, such as Matrix Profile-based approaches and our RS4 (Restricted Search Space for Snippet Selection), achieve strong efficiency and pattern fidelity, their adoption remains limited by usability barriers. Existing algorithms typically produce raw numerical outputs that require technical expertise to interpret, and few accessible interfaces support configuration or visualization. In this demo, we present GUISSE, an interactive graphical interface for executing and evaluating snippet discovery algorithms. The platform enables flexible experimentation with RS4 and baseline methods, allowing users to configure clustering techniques, distance weightings, and snippet parameters. By bridging algorithmic execution with interactive exploration, GUISSE empowers a broad range of users to apply, interpret, and validate snippet discovery results without programming expertise, promoting transparent and reproducible time series summarization workflows.

Palavras-chave: Snippet Discovery, Time Series Summarization, Interactive Visualization

Referências

Das, M. and Ghosh, S. K. (2018). Data-driven approaches for meteorological time series prediction: a comparative study of the state-of-the-art computational intelligence techniques. Pattern Recognition Letters, 105:155–164.

Fernandes, G., Gaspar, L. P., Cruz, L. A., Magalhães, R. P., and Macedo, J. A. (2025). Rs4: Restricted search space for snippet selection. In Intelligent Systems. Springer Nature Switzerland.

Imani, S., Madrid, F., Ding, W., Crouter, S., and Keogh, E. (2018). Matrix profile xiii: Time series snippets: A new primitive for time series data mining. In 2018 IEEE International Conference on Big Knowledge (ICBK), pages 382–389.

Law, S. S. and Cervone, D. (2021). Stumpy: A powerful and scalable library for time series data mining in python. Journal of Open Source Software, 6(60):2794.

Liu, X., Sun, H., Han, S., Han, S., Niu, S., Qin, W., Sun, P., and Song, D. (2022). A data mining research on office building energy pattern based on time-series energy consumption data. Energy and Buildings, 259:111888.

Löwe, S., Schmidl, S., Fambom, K., et al. (2022). sktime: A unified interface for machine learning with time series. arXiv preprint arXiv:2209.06844.

Yeh, C.-C. M., Zhu, Y., Ulanova, L., Begum, N., Ding, Y., Dau, H. A., Silva, D. F., Mueen, A., and Keogh, E. (2016). Matrix profile i: All pairs similarity joins for time series: A unifying view that includes motifs, discords and shapelets. In 2016 IEEE 16th International Conference on Data Mining (ICDM), pages 1317–1322.

Zhu, Y., Zimmerman, Z., Senobari, N. S., Yeh, C.-C. M., Funning, G., Mueen, A., and Keogh, E. (2016). Matrix profile ii: Exploiting a novel algorithm and gpus to break the one hundred million barrier for time series motifs and joins. In IEEE ICDM, pages 739–748.