Introduction

This demo website features two new and useful functions that are currently missing from the National Library 
of Medicine (NLM)’s Journals database. First, it displays journals in the same discipline by popularity 
(how frequently a journal is viewed by PubMed users). Second, the Related Journals link suggests journals 
closely related to a journal of your choice.

Usage Instructions

1) You can locate a journal in two separate ways: 

   a) Type a journal’s name, abbreviation, or ISSN in the query box and hit the “Search” button. All 
      matched journals are displayed if there is more than one journal matching the keywords. 

   b) Click on the “Subject Terms” link on the homepage and then click on the subject of your interest. 
      Journals are displayed in two different orders in the result page, by title or by popularity. 

2) You can locate related journals by clicking on the Related Journals link, which will take you to a 
   result page of 20 related journals, if available.

  We compute the network of related journals by mining journal co-occurrence patterns from the aggregated 
  PubMed log data (from millions of users). Given a journal A as input, we compute probabilities that a 
  user shifts from A to other journals and subsequently use those probabilities to rank the other journals. 
  The larger the probability, the closer another journal is to the given journal A. The 20 most closely 
  related journals are selected for output.

  A total of 15,827 journals are currently used for this demo website. They are made available because 
  they appeared in the PubMed log data we collected (March 2008). Our data include most of the journals 
  currently indexed in MEDLINE, as well as many others exist exclusively in PubMed. (see here for the 
  difference between MEDLINE and PubMed).
  

  Journals ranked higher by usage are more popular among PubMed users. We do not view the usage data as a 
  proxy for the prestige of a journal, nor its importance to its field. Instead, the popularity of a journal 
  as determined by its usage should be interpreted as the users’ choice. 

  There may be several reasons for this. First, the journal is currently not indexed for this demo website. 
  Second, you typed a different name than its indexed name. In this case, you can try one of the following: 
  1) type its abbreviation (if known) 2) type its ISSN or 3) browse for the journal by Subject Terms. 

  In our system, we allowed abbreviations as search keywords as long as they are indexed in the NLM's Journal 
  database. Unfortunately, some commonly known and widely used journal abbreviations (e.g. NAR) are not 
  currently indexed in the NLM's Journal database. Such a problem will be resolved when those abbreviations 
  get added to the Journals database. We will keep updating our Website on a regular basis.

  Before computing the related journals, we filtered robot sessions in order to stop noisy clickthrough data 
  (e.g. spam clicks) from affecting our results. Specifically, we applied two heuristic rules to identify and
  discard robot sessions. The first rule we employed checks the user-agent information from the user browser 
  against a list of known browser types (e.g. mozilla). We also excluded all sessions with more than 50 citation 
  retrievals or less than 2 retrievals, since most of our users are not likely to view more than 50 abstracts 
  in 24 hours (on average, 6 retrievals per user session) and single retrievals are not useful in our analysis. 
  After applying these filters, 8 million user sessions remained, totaling 51 million retrievals.

  The graphic bar, associated with a journal on the result page, measures the relative strength of popularity 
  or similarity of that journal compared to the highest ranked journal, whose scores (either popularity or
  similarity) of the top ranked journal are normalized to be one.

  We replaced retrievals of PubMed citations with their corresponding journal titles in this study. The number 
  of accesses to a given journal displayed in the search results (by subject terms) is equivalent to the number 
  of times papers of that journal were accessed through PubMed within a month. 

  Our database imports journal information from National Library of Medicine's Catalog database. Therefore, 
  please check the journal information at the NLM Catalog [http://www.ncbi.nlm.nih.gov/sites/entrez?db=nlmcatalog]
  and contact the NLM's Helpdesk if you need to request changes. Records in our website will be in sync with 
  the NLM's Catalog.

Citation

  Below is its abstract.

  ABSTRACT
  Motivation: With the explosion of biomedical literature and the evolution of online and open access,
  scientists are reading more articles from a wider variety of journals. Thus, the list of core journals 
  relevant to their research may be less obvious and may often change over time. To help researchers 
  quickly identify appropriate journals to read and publish in, we developed a Web application for finding 
  related journals based on the analysis of PubMed log data. 
  Availability: http://www.ncbi.nlm.nih.gov/CBBresearch/Wilbur/IRET/JOURNALS/

Contacts