A Quantitative Comparison of Distance Measures for Trajectories
Budget: €30 – €250 EUR
Take the http://zheng-kai.com/paper/vldbj_2019.pdf survey on distance measures that
can be used to evaluate the utility of anonymized trajectories vs
the original trajectories.
A Quantitative Comparison of Distance Measures for Trajectories
Basically, it reviews distances measures d(T_1,T_2) between a trajectory
T_1 and a trajectory T_2.
Then it would be a matter of implementing (maybe only a subset of)
those trajectory distances in python and comparing them in terms of:
- Computational cost.
- Correlation between the various distance measures when
applied on a large data set of trajectories. Do most distances
yield similar values? Are some distances that behave substantially
differently from the others?
In summary, the idea is to add the quantitative comparison
between distances
can be used to evaluate the utility of anonymized trajectories vs
the original trajectories.
A Quantitative Comparison of Distance Measures for Trajectories
Basically, it reviews distances measures d(T_1,T_2) between a trajectory
T_1 and a trajectory T_2.
Then it would be a matter of implementing (maybe only a subset of)
those trajectory distances in python and comparing them in terms of:
- Computational cost.
- Correlation between the various distance measures when
applied on a large data set of trajectories. Do most distances
yield similar values? Are some distances that behave substantially
differently from the others?
In summary, the idea is to add the quantitative comparison
between distances