I'm trying to figure out the best strategy for choosing what words to add that I come across. I regularly add words that I'm studying "formally" (from a book or ChinesePod), but in casual reading (news, novels) I find that I can scrape 50 or more words from a given article. Obviously its not realistic to bomb my deck with all the new words I come across, as in the end it just discourages me from reading instead (to avoid ruining my SRS schedule with leeches, or drowning out the formal lists that I want to keep a tight schedule on).

What I've done so far is try to implement some self-discipline, i.e. only choosing a couple words from a given article, or perhaps a word every few paragraphs in a book, but inevitably I end up adding words of little value because I can't necessarily judge their utility upon my first time seeing them.

What I'd like to do is export a list of the words I looked up in Pleco each day and copy them into some program which then sorts them according to a frequency list. Then I'd just add the 5-10 per day that I came across which I'm most likely to come across again. Is anyone aware of a online or offline app that does this?

Thanks!

The way a professor once explained reading in a non-native language to me was: "Skim over all the words you don't understand, and only look up the ones that pop up over and over again, otherwise you will drive yourself crazy."

That's a good rule of thumb for *looking up words*; it's essentially what I have been doing, but I think its inefficient as a method of selecting what words *to add to SRS lists*. Anyway, the fact that I'm reading in Pleco means that looking up any word is essentially 'free' (time-wise), so I need some filter between the words I look up and the words I add to a list to be studied. Obviously some of the ones I look up are worth adding, but its hard to judge that based on frequency just within a few pages/single article.

On another forum this suggestion was offered, which seems to meet my needs:
"You can paste your words in the form here: http://mandarinspot.com/annotate, check "For printing" checkbox, then select "Add vocabulary for all words" and "Sort by frequency", and click Annotate button. Not exactly the reason the tool was created, but should work for your case nonetheless. "

Another tool that I've found super useful is Jun Da's Vocab. Profiler, available at: http://lingua.mtsu.edu/chinese-computing/vp/index.php?CNTEXT_Session=fe5b47a79b954e9e0b088d6d97af10a8.

You can copy and paste any article, or even a whole bunch of them into his vocab. profiler and it will tell you the most commonly used characters and bigrams (even trigrams and beyond all the way to N-grams, but that's not mostly going to be very useful unless you're trying to sniff out chengyu or something). I think it's a good way to do essentially what Joshwhitson13's prof. suggested, and one of the best things about it is that you can determine the most commonly used words in a small selection in a specific area (for instance if you are interested in making a special vocab list) or in every piece of digital text you ever read.

Just incidentally, I also used this recently to help me search engine optimize a new Chinese language website I am creating, and it worked really well too!