Context: A lemma is the uninflected form of a word. So while “tree” and “trees” are two words, they are the same lemma: “tree”. Similarly, “go”, “went” and “going” are all forms of the underlying lemma “to go”. This dataset contains the most common lemmas in Japanese. Content: This dataset contains the most common Japanese lemmas from the Internet Corpus, as tagged by the ChaSen morphological tagger for Japanese (http://chasen.naist.jp/hiki/ChaSen/). For each lemma, both the frequency (number of times it occurs in the corpus) and its relative rank to other lemmas is provided. The total corpus size is 253,071,774 tokens, with a lexicon of 451,963 types. Acknowledgements: This dataset was developed at the University of Leeds by Centre for Translation Studies(more information: http://corpus.leeds.ac.uk/list.html), and is distributed under a CC-BY license. Inspiration: This dataset is an especially helpful resource for work on Japanese texts. * What is the distribution of hiragana, katakana and kanji characters among common lemmas? * Can you use machine translation to find the equivalent lemmas and their frequency in other languages? Is there a lot of cross-linguistic difference between what concepts are the most frequent? * Which parts of speech are the most common in Japanese? Are these different across languages?