[BLT] Blog for Library Technology is published by the University of Michigan Library's Library Information Technology division. We'll talk about technological innovations we're developing in the U-M library. We welcome your comments and feedback! (If you don't have a U-M uniqname, you can sign up for a free Friend account and log in with it to post comments.)
Advanced Search for HathiTrust full-text seach
April 18, 2012
In February we released the first part of the advanced search interface for HathiTrust full-text search. Advanced search allows users to combine a full-text search with searches within specific metadata fields such as Title, Author, or Subject.
For example if you want to find out where Charles Dickens used the phrase "the best of times" you can search for: [All of these words] [Dickens, Charles] in [Author] AND [This exact phrase][the best of times] in [Just Full Text]
The advanced search interface also allows you to set limits by publication date, format, or language. Multiple languages or formats can be selected.
Today we released the second phase of advanced search. You can now combine up to four different fields connected by the "AND" or "OR" operators, and any limits set are retained if you click on the "Revise this advanced search" on the search results page.
Posted by Tom Burton-West on April 18, 2012. Permalink | Comments (0)
Tagging and Favoriting
April 02, 2012
Since February 2008, the University Library has offered a service called "MTagger" as a way to allow site visitors to save resources for future use. (I wrote about the service several months after launch in an article titled "MTagger Update.") MTagger was patterned after Delicious, the popular social bookmarking site. The idea was that visitors to the library web site could save individual books, web pages, images from our digital library, and so forth for future use. People would "tag," or add their own descriptive keywords, to the items they saved. These tags would then be shown on the item, allowing future visitors to explore other resources.
MTagger Tag Cloud
In addition to public "tagging" of items, we also used MTagger as the storehouse for items "favorited" in the library catalog. A favorite is distinct from a tagged item in that favorites are personal and private, accessible only to the individual who saved them. For example, if I had "tagged" the catalog record for a particular book, my tag would have appeared for all users in the tag cloud. Anyone could have clicked it, seen what else might have been identically tagged (by me or by others), explored other items that I had tagged, or other tags that I had used. By contrast, an item that I marked as a favorite was available only to me. Using MTagger was an inherently public act; using favorites is inherently private.
Usage patterns over the four years that MTagger was part of our web site show a clear preference for "favoriting" items rather than tagging them. Of the total number of MTagger users and items, items saved through the catalog's favorites mechanism are the overwhelming majority (the "favorites" function in the catalog is new, existing for the most recent 2 1/2 years of MTagger's existence). MTagger's usage stats can be summarized as follows:
Use of MTagger & Favorites
| Measure | Total Number MTagger & Favorites |
Created through Favorites | Percentage From Favorites |
|---|---|---|---|
| Distinct users | 8,300 | > 7,000 | > 75% |
| Items tagged | 88,000 | > 72,000 | > 80% |
| Tags used | 9,300 | N/A | N/A |
Moving Toward Favorites
In fall 2011, the library launched "Search Tools Favorites", a way for authenticated library web site users to save databases and online journals (from Search Tools, our database and journal finder) and article citations (from ArticlesPlus, our SummonTM-powered article discovery tool). From its launch on November 2, 2011, through March 29, 2012, 549 library visitors have added a total of 4682 favorites (3097 article citations, 1012 databases, and 573 journals).
We are also in the process of migrating Mirlyn Favorites into the new Favorites system (they still are being saved as tags behind the scenes). Here, 305 users have favorited 3031 catalog items.
Our current Favorites tool is "siloed" -- that is, users who have marked items as a favorite can see them in separate lists in Search Tools, one each for articles, databases, and journals. Users who want to see their Mirlyn favorites must go to Mirlyn to see them. During the summer, we will be launching a new integrated favorites interface that will allow people to see all their favorites in one place and to organize them into categories -- so that books, articles, and databases for a single project can be listed together. We are still working on this interface and related programming. User studies for the interface will take place in the coming month, and we expect to launch the new integrated favorites interface in the first part of the summer.
Retiring MTagger
We are in the process of retiring MTagger. We have removed the MTagger tag cloud from the catalog and DLPS image collections. At some point in May, after Commencement, we will remove the tag cloud from the footer of pages on the library web site, as well. Where we can, we will migrate any items saved with MTagger (items from the catalog, databases, or journals) into favorites.
If you have questions, please use the Web Systems Feedback Form to reach us.
Posted by Ken Varnum on April 02, 2012. Permalink | Comments (0)
Bookmarks, Favorites, and Lists: A Question of Names
March 01, 2012
At the University of Michigan library, we are working on such a tool to integrate four separate tools (one for each of these kinds of resource: databases, online journals, library catalog records, and article citations). But what should we call such a tool, one that is designed to allow site visitors to save resources into their account for later access?
Several weeks ago, I sent a link to a quick poll to a couple of listservs looking for information about what libraries have chosen to name their "save this for later" or "favorite" tool on the site. I received 12 responses (actually, 13; two people responded independently on behalf of Northwestern University's version of the tool; I consolidated their responses into one). The following table summarizes the information we received.
| % | Name | Number | Satisfied | Notes |
|---|---|---|---|---|
| 23.1% | Bookmarks | 3 | 3 | |
| 15.4% | My Favorites | 2 | 2 | Drexel's implementation |
| 7.7% | Book Bag | 1 | 1 | |
| 7.7% | Cart | 1 | 0 | Makes me feel I'm buying something on our catalog |
| 7.7% | GalterLists | 1 | 1 | Northwestern University's implementation |
| 7.7% | My Library | 1 | 1 | "My Library" encompasses a bunch of stuff including "Saved Databases," "Saved Journals" and "Saved Citations" plus all the recommendations we make for the three categories. |
| 7.7% | My List | 1 | 0 | Its function is not as clear as I'd like, but fit within the space constraints of the tab and was the most self-explanatory of the options we looked at. |
| 7.7% | Other | 1 | 1 | Had 'favorourites' but it confused patrons (mixed up with IE's bookmarks) |
| 7.7% | Save for Later | 1 | 1 | |
| 7.7% | Shelves & Lists | 1 | 1 | Shelves are public; lists are private. Boston Public Library's implementation |
"Bookmarks" and "My Favorites" were the most common answers (5 out of 13 responses), and were the only names that garnered more than one response. Three respondents had some flavor of "list" (GalterLists, My List, and Shelves & Lists).
Most respondents reported being satisfied with the name they selected.
So what are we going to name our implementation? We're leaning toward keeping the same name as we have now, "Favorites," despite a small amount of discomfort with the idea that a scholarly article or a book is really a "favorite" resource, when it is simply being saved for future use. It's not perfect, but its function seems to be broadly understandable to users who favorite things on various web sites all the time, from Tweets to URLs in Internet Explorer.
Posted by Ken Varnum on March 01, 2012. Permalink | Comments (0)
Wikipedia Blackout and MLibrary Site Search
January 20, 2012
The Great Wikipedia Protest Blackout of 2012 did not result in any particularly significant increase in site searching at the University of Michigan Library. While traffic was up on January 18, 2012, over the same day the previous week, (January 11), the increase was about the same as for the day before and the day after -- reflecting the increasing workload of the academic semester more than any Wikipedia-inspired bump.
January 18 compared to January 11, 2012
Here are some numbers to illustrate the point. For "Outage Wednesday" compared to the previous Wednesday (January 11), searches were up slightly: 4% overall, and 14.5% for the default default "MLibrary" site search:
| Search Kind | 1/18/12 | 1/11/12 | Change | Percent |
|---|---|---|---|---|
| MLibrary | 4260 | 3719 | 541 | 14.55% |
| Catalog | 4518 | 4870 | -352 | -7.23% |
| Articles | 1805 | 1579 | 226 | 14.31% |
| Total | 10583 | 10168 | 415 | 4.08% |
January 17 compared to January 10, 2012
However, a somewhat larger overall increase is noted between the Tuesday before (January 17) and the Tuesday a week earlier (January 10): up 9% overall and 10% for the default "MLibrary" site search:
| Search Kind | 1/17/12 | 1/10/12 | Change | Percent |
|---|---|---|---|---|
| MLibrary | 4297 | 3894 | 403 | 10.35% |
| Catalog | 5326 | 5427 | -101 | -1.86% |
| Articles | 2211 | 1465 | 746 | 50.92% |
| Total | 11834 | 10786 | 1048 | 9.72% |
January 6-12 compared to January 13-19, 2012
For a 7 day week extending from the Friday before the outage to the Thursday after, when compared to the previous week, we see an actual decrease in Outage Week over the week before (a decrease of 2.45% overall, although a small 2.2 percent increase in the default "MLibrary" site search:
| Search Kind | 1/13-1/19 | 1/6-1/12 | Change | Percent |
|---|---|---|---|---|
| MLibrary | 23640 | 23130 | 510 | 2.20% |
| Catalog | 28499 | 32906 | -4407 | -13.39% |
| Articles | 11114 | 8805 | 2309 | 26.22% |
| Total | 63253 | 64841 | -1588 | -2.45% |
Posted by Ken Varnum on January 20, 2012. Permalink | Comments (0)
HathiTrust "Search in this text." Now with relevance ranking and better multilingual support!
August 29, 2011
Today we released the third high priority feature identified by the HathiTrust Full-Text Working Group: Relevance ranking for "Search in this text." Now when using the "Search in this text" feature, instead of having to scroll through numerous pages of results in page order, the results are now returned in relevance order with most relevant pages at the top of the list. The default is to list only pages that contain all the words in a user's search (a Boolean "AND" search.) However, there is also a link that will search for pages containing one or more of the search terms. If this option is selected, the pages containing more of the user's search terms are ranked higher.
In addition to relevance ranking, searching for non-Latin languages such as Hindi, Arabic, Hebrew, or Thai, now matches the capabilities of the Full text search of all 9 million volumes.
Posted by Tom Burton-West on August 29, 2011. Permalink | Comments (0)
HathiTrust Full-Text search: Now with Facets!
August 10, 2011
On July 27th we went live with faceted search and relevance ranking based on both OCR and MARC metadata in Full-Text search. (www.hathitrust.org) These are the top two features identified by the HathiTrust Full-Text Working Group.
http://www.hathitrust.org/full-text-search-features-and-analysis
The relevance ranking now will give volumes that match a user's query terms in both the OCR and in the title or author or subject a higher ranking than a match in only the OCR. There is much more work to be done in tuning relevance ranking, but this is a first step.
Search results can now be refined by selecting facets such as subject, date or author. Although selecting facets can help users drill down to narrow large result sets, using very specific terms and especially using phrases in quotes remain one of the best ways to get reasonably small result sets.
Over the next few months we will be releasing further improvements in ranking and more of the features identified by the task force.
Posted by Tom Burton-West on August 10, 2011. Permalink | Comments (0)
Updated Site Search Results
July 22, 2011
Several months ago, the library’s Usability Working Group and User Experience Department completed a review of the way large academic libraries present combined search results on their web site. (See "A Look at Combined Search"). That review was the first phase in a process to help us revise the way we presented our combined search results with an end goal of deciding whether or not to include results from our article discovery environment in the mix with everything else.
As we looked at what our colleagues were doing and solicited feedback from library staff and library patrons, we realized that before we could contemplate adding another chunk of information to our search results, we needed to streamline what we already have. Therefore, we embarked on a redesign of the interface to give us room for new content and to improve the usability of what was being shown.
Google Analytics told us that most clicks to results were on the first 2 or 3 items in a result list; in some categories (such as our catalog) we were presenting up to 10 results. User feedback told us that we were not providing enough context for individual results to tell the user what they would get when they clicked. And just about everyone we asked noted that the results were presented in too small a font size for comfortable reading.
In response to this range of data and comments, we have revised the interface, giving more space to search results and drawing more attention to other sources of information for our site’s visitors. We have reduced the number of items in each section to a maximum of three, with more prominent "see all" links to get to all results for each category. (See screen shots at the bottom of this post -- or try it yourself from the library site.
We continue to present librarian subject specialists to match the search query in the right column, but have pulled out staff directory contacts to that column as well, removing them from the library web pages section. We have also added a new “Didn’t Find It?” section to the bottom center of the search results, inside of which are links to conduct the same search in ArticlesPlus (our Summon article discovery tool), Deep Blue (our institutional repository), and the HathiTrust. Links to subject-focused browse pages are also included here.
This redesign has put off the question of including ArticlesPlus results in the search results page, although we are likely to include them here now that there is space to do so. We are leaning this way for two reasons. First, because a review of search queries indicates that about a quarter of the searches conducted on our site would benefit from the inclusion of article-level results. Second, because of direct patron and staff queries requesting this feature.
New Search Results
Click for a full-size image.
Original Search Results
Click for a full-size image.
Original Search Results
Posted by Ken Varnum on July 22, 2011. Permalink | Comments (0)
Announcing a Drupal Module for Searching Summon via API
June 03, 2011
The University of Michigan Library's Web Systems department built a Drupal module for searching Serials Solutions' Summon product using the Summon API. Developed by Web System's lead developer, Albert Bertram, the Article Discovery module conducts searches in Summon, replicating much of the search functionality offered by Summon's native interface. It presents search results with facets, offers an advanced search, and allows users to export selected records to EndNote, RefWorks, or a simple citation list that can be emailed.
You can see the Article Discovery module in action on the University of Michigan Library web site. Simply type your query in the search box and click "Go."
The Article Discovery module requires a Summon API key, Drupal 6, PHP version 5.x, and a copy of the Summon API library for PHP to operate. The module provides a configuration page, a search results page, and blocks for displaying a search box and a facet box. Though the Article Discovery module will work in a generic Drupal theme, you will likely need some customization to fit cleanly into your library's existing site.
The Article Discovery module is available for installation: http://drupal.org/sandbox/bertrama/1119778. We look forward to hearing your experiences when using this module. Please provide feedback through the Drupal site or by contacting us directly.
Posted by Ken Varnum on June 03, 2011. Permalink | Comments (0)
A Look at Combined Search
May 12, 2011
The Usability Group and User Experience Department have partnered on a project to improve the display of library website search results. A search of the library's website using the default "MLibrary" tab currently retrieves:
- databases
- items listed in the Mirlyn catalog
- online journals
- research guides
- webpages from the library's website
- collections (both digital and other collections)
- items from the library's database of government documents
- items listed in the UM institutional repository, Deep Blue
- relevant library contacts, such as specific subject specialists and services
In each category, a maximum of 4-10 matches are displayed, depending on the category and the number of matches found.
Continue reading "A Look at Combined Search"
Posted by Julie Piacentine on May 12, 2011. Permalink | Comments (0)
NIH/NLM Oral Histories Served by DLXS
December 07, 2010
As the creators of DLXS, we're quite impressed with their adaptation of Text Class by NLM/NIH to provide access to oral histories. The following is an excerpt from the press release.
The National Library of Medicine's History of Medicine Division is pleased to announce the release of a new Web interface (http://www.nlm.nih.gov/hmd/manuscripts/oh.html) to its oral history collections, as part of its growing electronic texts program. Content includes digital editions of transcripts and any accompanying audio content when feasible. Users can browse content by title, interviewee name, and subject. Full-text searching is available across all sub-collections, across each sub-collection, and within each transcript.
Browse the Archives and Modern Manuscripts Oral Histories to jump right into the audio content.
Posted by John Weise on December 07, 2010. Permalink | Comments (0)
ArticlesPlus Browser Bookmarklet
November 19, 2010
Ever visited a web page and wanted to find more information, from the academic or popular press, on the same subject? This bookmarklet will automatically search ArticlesPlus for the title of the page you are visiting. (You can edit the query if you like, to remove or add other keywords.)
What It Does
Say, for example, you're on the Wikipedia page for President Franklin Roosevelt. You want to quickly find scholarly articles about him. Click the bookmarklet and you're given a box to edit the search query:
Click "OK" and you're taken to an ArticlesPlus search results page for that same query:
Where to Get It
The ArticlesPlus bookmarklet is on the University Library's MLibrary Labs site.
Credit
This bookmarklet is based on a similar one developed by Barbara Arnett and Valerie Forrestal at the Stevens Institute of Technology's S.C. Williams Library.
Posted by Ken Varnum on November 19, 2010. Permalink | Comments (2)
Mirlyn Mobile
November 11, 2010
Have you ever wanted to look up a book when you were deep in the heart of the library stacks, riding the bus, or between sets at the gym? We have! The User Experience Department, along with Library Systems, is proud to present Mirlyn Mobile. Mirlyn Mobile is designed to bring the powers of the library catalog to your handheld device. Take your research needs in stride with the smartphone-optimized mobile catalog.
To access Mirlyn Mobile, point your mobile device's web browser to m.mirlyn.lib.umich.edu (works with desktop browsers, too).
Posted by Suzanne Chapman on November 11, 2010. Permalink | Comments (0)
ArticlesPlus: Article Discovery on MLibrary Site
September 27, 2010
Our Articles search is now even better. The University of Michigan Library is now using Serials Solutions Summon article discovery tool for the new improved ArticlesPlus tool. Now, when you search the ArticlesPlus tab on the library web site, you are searching across 190 million full-text and citation-only records available to the University of Michigan community through the University of Michigan Library.
Rather than using either Summon's default interface or VuFind (which we use for Mirlyn, the library's catalog), we built the tool in the library's web site in Drupal. We did this using a custom module that interacts with Summon's Solr index through their API to retrieve search results and facets. (If you are interested in the Drupal module we developed, please contact us -- we're putting together a version for public release later this fall.) The site's design was created by the library's User Experience Department working with a group of librarians.
Just as with any Summon installation, anyone can search ArticlesPlus and see citations. It is at the point of clicking through to the full text by clicking the "MGet It" icon, taking you through to our link resolver, that you must authenticate.
In the next few months, we'll be adding articles searches to our one-search MLibrary search and adding librarians to the articles results pages. We're also soliciting feedback from ArticlesPlus users through a survey link on the site. If you try ArticlesPlus, let us know what you think!
Posted by Ken Varnum on September 27, 2010. Permalink | Comments (1)
New List of Digital Collections
September 17, 2010
The Digital Library Production Service or DLPS (part of our MLibrary Library Information Technology division) has created and hosted a very large number of digital collections over the last 10+ years. We have been working for many years to integrate those collections into MLibrary services, and we are now ready to present the next link in this chain-- a more easily navigable and more fully featured list of these collections:
http://quod.lib.umich.edu/lib/colllist/
The road to building this collection list has been long and sometimes difficult. Four years ago, Suzanne Chapman and John Weise discovered the Exhibit tool (part of the SIMILE set of tools from MIT Libraries). They worked through many of the initial hurdles regarding how, where, and what we host using this tool. After a languishing period, it was revitalized by Kat Hagedorn, Jose Blanco, Roger Espinosa and Saurabh Koparkar to its current state. We would not have gotten half so far without two ULAs who kindly volunteered their services to help categorize, describe and find image thumbnails for each collection-- Ellen Wilson and Lorelei Rutledge. Many hearty thanks to all involved in the process of putting this tool together.
The new collection list replaces the the old, which was created originally to gather administrative information about each collection, including the responsible party, and statistics about size and usage. Public discovery and usage of the list was not intended nor anticipated, but the list became a popular public access point for DLPS collections. The old list can still be seen: http://quod.lib.umich.edu/cgi/c/collsize/collsize. Better, no?
Posted by Kat Hagedorn on September 17, 2010. Permalink | Comments (0)
Putting a Librarian's Face on Search
July 26, 2010
When you do a search on the University of Michigan Library's web site, you get not only results from the catalog, web site, online journal and database collections, and more, you also get a librarian who is a subject specialist related to your search term. While the matching is not perfect, it provides a human face on search results. So, for example, if you search for "Kant," in addition to books and databases, you also get the subject specialist librarians for humanities and philosophy. A search for "Jupiter," you get the subject specialists for Astronomy & Astrophysics and Humanities (after all, we don't know if you're searching for the planet or the Roman deity). When we can't make a reliable match to a subject specialist, we provide a link to Ask a Librarian, our reference service.
How does the matching work?
The University of Michigan library has long maintained a database of Library of Congress Call Numbers and "Academic Disciplines" -- which is what we call our subject taxonomy. (You can see it in action in the site's Browse function.) These categories broadly mirror the schools and departments at the University of Michigan. Librarians have assigned Library of Congress call numbers to each academic discipline. This mapping was originally done for our New Books tool so that students and faculty could find out when a new book related to their area of study was acquired by the library. A single call number can be assigned to multiple Academic Disciplines, so a given book could appear in multiple places.
In a site search, we do a special query of the library catalog behind the scenes and get the first 100 catalog results (sorted according to the catalog's relevance ranking algorithms). We sort those results into Academic Disciplines. If more than 25 items are in a single Academic Discipline, we include the subject specialist responsible for that particular area. (We set the threshold at 25 matches to help ensure a relevant match, but a librarian specializing in the "wrong" subject is arguably better than no librarian at all.)
We make the call number-to-academic discipline mapping available on our site at http://www.lib.umich.edu/browse/categories/. There is also an XML version of the mapping free for all to use or adapt.
Posted by Ken Varnum on July 26, 2010. Permalink | Comments (3)
See the full archives for this blog.
