Install the npm libraries webpack. For any other extra js libraries
add the package name to package.json and require the package in the
background.js.
When testing, after saving changes run:
npm install
npm run build
before reloading in the browser
- color common words
- select + key-combo to save off word in page (word, url, context)
- highlight words from history in green
- Do similar frequency analysis and shading for N-grams (N>1)
- How many words to highlight (not hardcode 100)
- extend to Hungarian/English/...
- should program target specific languages OR treat all text equally? (i.e. could throw out words not recognized in target language to avoid confusing the analysis)
- Shaded color based on word frequency (darker more frequent)
- how to find main text on a page not wikipedia (white-list of common main text div/classes, ask user for div/id tag on given page, maybe based on selection,..)
- Part-Of-Tag tag coloring
- Ability to change between coloring schemes (freq, POS,)
- Bokeh pop-up graphs
- hover word popup
- rank out of all words on page
- number of occurences
- add storage to save vocabulary (save word + context, urla of the word)
- color scheme: frequency color according to some larger corpus (subtlex-it corpus maybe)
- export stats table in html file
- color scheme: frequency color according to saved vocabulary frequency
- color scheme: N-grams collocations frequency
- user interface for change vocabulary or history