Next we will learn The fourth view way, 'compare' After selecting compare There is an obvious change happened on the interface. This is to comparision between two objections. For example, we want to see The difference between "fast" and "quick". In the word search box, enter both words "fast quick". We may be more concerned what is the noun they directly used in the collocation area of collocate, Enter [nn *]* represents any terms behind the term behind the word "fast" or "quick". It appears after it. Of course, we can also continues to set the following content. For example, set a specific type of example for it, sort by possible years, and so on. This is familiar with what displayed under "list". Only now is the entry into force of the two search terms We donnot set it one by one. Click "search". The right part displayed the search results. The first table is "fast", The second table for "quick". First, here's the first digital watch Here is the fast: 1.21 representatives frequency ratio of these two words "fast" and "quick" in the entire corpus. The frequency of "fast", is 1.21 times of that of "quick". Take a countdown, quick is 0.83 of fast. See the first line of the table fast track The first column represents the frequency ratio of fast track. There are fast food fast lane, etc. Click it. See below in the quick view The second column refers to the quick track quick food quick lane You can see the frequency ratio appeared is small. The third column is the ratio of the first two columns score score is through The third column is divided by this value 1.21. It is a word exceeding the average gap The greater the value of the word, the more chance to be modified by fast. Can not be modified with quick is vice versa. On the right table of "quick", we can check the words which modify "quick" often, and cannot modify "fast". Just see the default sort is the relevance Sort by MI values If we change it to sort by frequency then click search You can see the two tables on the right side Actually respectively w1 and w2 in accordance with Sort of word frequency count The difference is just now We see the different colors. There are two types of red and green and different shades We observed a little difficult to find Dark green represents w1 is larger than w2. That is the word used w1 modified This is basically no danger. For example, we can say that fast food, instead of quick food. Green is actually lighter, shows that the ratio of w1 and w2 not as big as the original one. If you become a nearly white, there is no much difference between these two words. If the value in the table w1 is exceeded by w2, it represents that the words rank top in w1 is less often modifed by words in w2. Here it will be displayed in red The situation like this will hapeen when we'll see pull down. So when we want to compare two words simplely, differences in the collocations, we often use the default sort order. Set "relevance", here in the sort option. Notice, comparing two words like this, is not necessarily use compare. You may have noticed sections Here obviously there are two menus. We never used the second one. Show you now. For example, I want to know which words can be described by the word "hard", in corpus in magazines and in the academic corpus. Let's reset then in the word area enter the hard spaces [nn *] and select sections The first one we chose 'magazine' The second we chose the 'academic' then click search See the right side of both tables records the usage of word "hard" in magazines corpus and in the academic corpus, and the most possible word to describe. Similarly, If I want to see noun modified by "green" in 1990-1999 and 2000-2009/ What changes occurred between the two decades? I typed in the word box, green input box [nn *] Then in this section Choose 2000-2004 and hold down the ctrl, multiple choice. Then click 2005-2009 similarly, in the second menu select the 1990-1994 and 1995-1999, then click search to view its results. Okey, this is brief introduction of search method of BYU corpus above. More set solutions are using. You can click at any moment, the question mark on the interface, to learn more content.