Skip to main content
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: IEEE Trans Knowl Data Eng. 2018 Mar 5;30(10):1825–1837. doi: 10.1109/TKDE.2018.2812203

Fig. 10.

Fig. 10.

Efficiency evaluation of AutoPhrase on the three largest datasets. Both the running time and the peak memory are linear to the corpus size. Because of an almost lock-free parallelized implementation, the multi-threading speedup is close to linear.