transbigdata-preprocess

Clean noisy GPS trajectory data by applying boundary filtering, polygon clipping, and sampling interval analysis to identify missing records and out-of-bounds points, then use ID reindexing to separate interrupted trips into distinct trajectories.

ID reindexing separates long interruptions or spatial jumps in GPS logs into distinct trajectories by applying configurable time and distance thresholds, such as a 3600-second gap, ensuring continuous trip segments are correctly identified for downstream analysis.

Filter GPS records to a specific study area by applying bounding-box and polygon-based clipping to remove out-of-shape data points, improving spatial data integrity and ensuring trajectory records align with the target geographical boundary.

Analyze sampling intervals in transportation spatiotemporal data to surface sampling irregularities, calculate sample durations, and generate time range summaries that reveal missing data points within raw GPS datasets.

Preprocess taxi GPS logs for OD extraction by running a data quality summary, filtering records to the study city boundary, reindexing interrupted trips with a time gap threshold, and exporting the cleaned dataset.

Yes, transbigdata-preprocess works with raw GPS CSV files by performing data summaries, applying spatial filtering, executing ID reindexing based on time or distance gaps, and saving the cleaned result as a new CSV file.