New database with frequency of words

[mandarinboy]

If anyone is interested in it i am now parsing around 1.000.000 words a day from Chinese newspapers of different sorts to build a database with the most requently used words in modern Chinese newspapers. I will put it on my site http://www.mandarinteacher.com as soon as i am back from India.

The intention of the database is to help students learn commonly used words in todays newspapers. Later on i will also calculate texts with word density to make it possible to find texts that match the level of words a student do know.

What i have found is that the speach paterns varries widely betwwen for example peoples daily and 163.com. Maybe not so surprisingly but interesting.

Since the data i am using are from Cedict and other sources the database will be free of cruse but licensed under the same conditions as the Cedict data are today.

[in_lab]

If you are just checking for word frequency, why not use a more comprehensive word list? There are large word lists freely available, but you wouldn't have definitions for them.

[mandarinboy]

We are in fact useing wordslists without translations when parsing. This so that we get all words. What we are using cedict and adso for is to later on create lists for stundents with pinyin and translations. After all, the most frequently used words are in those lists so it is nice combinations.