Many users are unfamiliar with the various time series distance measures and are unsure which ones are most suitable for their specific tasks. To address this issue, it would be beneficial to provide recommendations on the appropriate distance measures for different domains and data properties.
Proposed Solution:
Domain-Based Recommendations: Provide guidance on which distance measures are suitable for specific domains. For example:
Finance: Use distance measure X.
Neuroscience: Use distance measure Y for specific tasks.
Data/task Property-Based Recommendations: Offer recommendations based on the properties of the user's data. For example:
Temporally Shifted Data: Suggest elastic measures over lockstep measures.
Large Datasets with Speed Requirements: Recommend lower bounding measures or SBD (cross-correlations) for faster processing.
By implementing these recommendations, we can help users select the most appropriate distance measures for their time series analysis, improving the accuracy and efficiency of their work.
Many users are unfamiliar with the various time series distance measures and are unsure which ones are most suitable for their specific tasks. To address this issue, it would be beneficial to provide recommendations on the appropriate distance measures for different domains and data properties.
Proposed Solution:
Domain-Based Recommendations: Provide guidance on which distance measures are suitable for specific domains. For example:
Data/task Property-Based Recommendations: Offer recommendations based on the properties of the user's data. For example:
By implementing these recommendations, we can help users select the most appropriate distance measures for their time series analysis, improving the accuracy and efficiency of their work.