```xml <result> <column_name>duration_ms</column_name> <family>additive</family> <subtype>ordinal_additive</subtype> <confidence>0.85</confidence> <requires_user_confirmation>false</requires_user_confirmation> <sum_meaningful>true</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>sum</primary_reducer> <valid_reducers>sum, mean, min, max, median, count, distribution</valid_reducers> <excluded_reducers>product (not a ratio), geomean (not a ratio)</excluded_reducers> <average_semantics>arithmetic_mean</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>min</inner_measure_reducer> <outer_measure_reducer>max</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>true</multiplicative_partner_derivable> <reasoning> Table type is PATH with sequence axis, making ordinal_additive plausible. Column name "duration_ms" with unit "ms" (milliseconds) is a physical extensive quantity (time duration). The statistics show large magnitude (max 6.5M), unbounded upper range, strictly non-negative values with zeros present, and center far from 1.0 (mean 24434, median 545) - all strongly additive signals. However, this is an instantaneous duration measurement at each sequence step on a PATH table, not a cumulative total. Summing durations across timestamps is ordinally meaningful (total time spent) but the values themselves represent point-in-time measurements. The high log_symmetry_gain (19.45) suggests right-skewed distribution typical of duration data. Classification: ordinal_additive because it's a physical quantity on a PATH table where cross-entity sum is meaningful but sequence-wise aggregation requires normalization. </reasoning> <table_type_conflict>false</table_type_conflict> </result> ```