This study explores the effectiveness of Data-Driven Learning (DDL) approach to second-language (L2) English vocabulary learning in on-demand online distance learning at a private university in Tokyo, Japan. The participants were 49 Japanese undergraduates, intermediate L2 English learners at the B1 level in the Common European Framework of Reference for languages (CEFR). The experimental group consisted of 25 students who completed verb-noun collocation correction tasks between the pre-test, and the post- and delayed post-tests. The control group consisted of 24 students who took the tests but did not perform the error correction task. The results show that for memorizing verb-noun collocations, on-demand online DDL -- in which learners refer to screenshots of concordance lines -- is as effective as DDL in a traditional on-campus setting, where learners search the Corpus of Contemporary American English (COCA) for themselves, even though the learning activities in remote online on-demand and face-to-face on-campus DDL are not exactly the same. This study suggests the potential of DDL in online on-demand classes. When a variety of teaching methods are being explored, DDL seems a viable option as an approach to L2 teaching. [For the complete volume, "Intelligent CALL, Granular Systems and Learner Data: Short Papers from EUROCALL 2022 (30th, Reykjavik, Iceland, August 17-19, 2022)," see ED624779.]