CAN'T DO COMPARE WORDS ON 1-2 LETTER WORDS; CREATES PROBLEMS WITH SCRIPT

不能比

只好手动比较了

The word in occurs 16,541,741 times in the corpus.

In COCA, you can usually find the collocates for high-frequency words like this, as long as:

1) you search by "lemma" (capitalize the "dictionary form" of the word(s), e.g. DECIDE instead of decides)
2) you leave the "span" set to 4 words left and 4 words right, and
3) you don't limit by section (e.g. by time period, genre, or dialect, depending on the corpus).
4) you don't use Virtual Corpora
5) you don't choose to see the frequency by section (i.e. you need to de-select the box to the left of "Sections" in the search form).

You might also re-do the search with one of the following:

1. Reverse the WORD and COLLOCATES fields, by putting the least frequent word in the [WORD] field, or
2. Do the search as a string search, e.g. ADJ health, instead of health + ADJ collocates, or VERB the money instead of money + NOUN collocates.
 

Most importantly, there is a MUCH better way to get lots of collocates for high frequency words, than searching for them one by one in the online corpus. WWW.COLLOCATES.INFO allows you to download millions of node / collocate pairs from either COCA (13.5 million node/collocate pairs) or iWeb (33 million node/collocate pairs). It will probably be much better for you, and it will definitely decrease the load on the corpus server as well.

单独搜IN的Collocates  prep词性的结果是空 调MIN MI 0/0.1 也不行。 其它仨词性有结果 ADV   CONJ   ADJ（ON的prep词性搭配也是这样； at有结果

没办法了，只能用cluster处理了，有点麻烦

https://www.collocates.info/files.asp

There are no entries for most of the 30-40 most frequent words (e.g. "the", "of", "with"), but all other words in the top 60,000 are included in the collocates data.This makes sense, at list from a simplistic psycholinguistic point of view. When most people hear bread, they think {loaf, slice, crumb, butter, eat}, etc. But it's not clear at all what they might think of when they hear the or of or with.