AISearch Firefox Plugin

AISearch Firefox Plugin

We present the AIsearch Firefox add-on which performs an on-the-fly categorization of search results obtained when searching the Web with Google. The add-on marks a first step toward making browsers intelligent tools to access the Web since the category formation is performed entirely within the browser. As a basis for the add-on we have developed the first JavaScript library for information retrieval which includes a selection of well-known retrieval models, clustering algorithms, cluster validity measures, and topic identification algorithms.


Save the XPI file on your desktop and then drag it into an open Firefox window. A pop-up window will appear which asks you to confirm the installation of the AIsearch Firefox Add-on. After the installation, click on the button “Restart Firefox”.


Open a Firefox window an visit Important: It has to be the English Google, not the German one! If you see “Deutschland” below the Google logo or if the Web page is localized to German click on the link “ in English” which is found in the lower right corner. Issue any search query you like, e.g. “tea flavor”, and then click on “Show categories”. This triggers the AIsearch retrieval process and after a short while the categories will show up.

Data security

The add-on does not communicate with any external service!

Known Issues

Don’t worry, the add-on will never crash your browser, however, there are some issues. In particular, the integration with Google is not yet perfect:

  • There may be interferences with native Google search options. Please disable all the search options.
  • On slower hardware Firefox may complain that the AIsearch script is not responding because the clustering takes longer than expected.

If you have signed in with your Google account:

  • Clicking on a category appends the snippets instead of replacing them.
  • Sometimes the categories won’t show up although the add-on says they are loading (wait max 10s for them).

Leave a Comment

Previous post:

Next post: