Support Options

Submit a Support Ticket


Wish List - Wish #86


Member avatar

0 Dislike

Nahil Sobh

Better Search Engine

Comments (9)

  1. Michael McLennan

    When you search for items like “ECE 453”, you often get individual lectures, instead of the top-level course that summarizes them. Also, items with 0 ranking (due to being new or a bug in stats processing) are ignored. We have no visibility into files like PDF, PPT, DOC, etc. There should be a link for “Advanced Search” and better instructions about how to narrow down to authors, seminars, etc. And it would be nice to have spelling corrections or other search suggestions. At the same time, we must be careful to guard our private content; those items can’t show up in search results unless the person searching is in the group that allows access.

    Reply Report abuse

  2. Gerhard Klimeck

    this is certainly a multi-stage process. 1) fix the ranking – there are many resources with ZERO ranking 2) the study of a reputation system and how this will influence searches

    3) topics pages must show up on the very top, when relevant

    Reply Report abuse

  3. George A. Howlett

    Examples would really help. I know it may seem obvious what the results *should* be. But it really crystallizes things when you can say

    here’s the query.

    here’s the actual search results.

    here’s what I would have expected to find.

    I echo the comment that “topics” should be ranked higher. Is this related to ranking titles over free text?


    Reply Report abuse

  4. Nahil Sobh

    I would recommend using Google Custom Search:

    Reply Report abuse

  5. Nahil Sobh

    Here is our implementation of the Google Search at our web-site:

    Reply Report abuse

  6. Alissa Nedossekina

    As a note: just enabled “Site Search Tracking” in Google Analytics. This should allow us to get reports on how our internal search engine is being used – what the common search terms are and where people end up going on the site. Hopefully it’ll give us some valuable info to improve our own search code. Generally, I think we are not using Google Analytics well enough. The application is loaded with features we aren’t using and we can benefit from. Of course, it’d take someone to look closely into this and do the human analytics part.

    Reply Report abuse

  7. Ashlie Martini

    Although I have no idea how to implement this wish…it is absolutely the most important item on the list from my perspective.

    Reply Report abuse

  8. George B. Adams III

    For the Google Analytics data should be made available to Diane Beaudoin, NCN Director of Assessment.

    Reply Report abuse

  9. Steven Snyder

    I’ve added a new search engine that addresses a lot of the specific complaints with search results, and hopefully some of the general issues with search results.

    I’m marking this wish as granted, though I understand that improving the search is an ongoing process, and if there is a finish line we’re doubtless not yet at it.

    The primary benefit of the new search engine is that it provides us with a way to respond much more quickly to complaints about bad search results for a given term set. My hope is that eventually, with enough such complaints responded to, the modifications we make along the way will lead to a general system that serves up results correctly for every set of terms.

    I’ve addressed some specifics already, namely increasing topic weights over that of other types of resources, improving the display of nested results such as courses by grouping them under a common banner, and extending the search to draw on tags to include things in the site search that were previously only found using the tag search.

    So, I hope the search is better. If you notice any problems please enter either a wish or a support ticket including the search terms, and how the results differ from your ideal.

    Reply Report abuse, a resource for nanoscience and nanotechnology, is supported by the National Science Foundation and other funding agencies. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.