1findr data is now continuously updated. New articles will be available via the search engine within minutes of when our acquisitions engine discovers them. Total number of articles climb throughout the day.
Minor modification of the deduplication algorithm
More than 1M articles from Chinese journals
Includes data harvested through October 8 – 100.7 million articles!
Addition of 1100 Chinese journals, roughly 300 other manually validated journals
Batch link revalidation complete, 18M more links have been revalidated
Complete Unpaywall data ingestion
Addition of 400 Taiwanese journals
Journal titles clean up
Fix of errors in the taxonomy
Merge around 90 journal duplicates
Determine OA color of nearly 2.4M articles
New interactive year-of-publication visualizationKeywords are highlighted in results (currently only when title & abstract have no highlighting)
Authors highlighted when searching via the author filter
1foldr Hub interface for researchers to claim their articles on their ORCID records
Accessibility: better use of lang tags, fix some ARIA labels
Partially fix bug where very (very) long queries would produce errors
PlumX metrics are inputs into the relevance formula
New sort options: most cited & most Mendeley readers
Removal of sources that were no longer updated
Removal of malformed data
Added 2160 journals
Removed 408 journals
Subject taxonomy updated
New subject taxonomy classifies an additional 5500 journals
Fix of an OA colours bug
Removal of sources that were no longer updated
Removal of malformed data
Added 2160 journals
Removed 408 journals
Subject taxonomy updated
New subject taxonomy classifies an additional 5500 journals
Fix of an OA colours bug
New data: 98.7 million articles
Pagination now fits better on mobile devices
Bugfix where page numbers weren’t obeyed when reloading / following a URL
Remove “index” endpoint for efficiency; translate subjects in browser
Fix URLs in exports
Fix issue with reloads and Publisher Country filter.
Approximately 1000 journals added!
Significant improvements to deduplication algorithm
Added ability to control access via profiles to the OA Link Resolver Service
Fixed a bug where full-text links failed on the article detail page
Added Open Access resolver endpoint for use by library link resolver systems
Fixed issue with autocomplete (journals, organizations, etc.) when search queries include a stopword
Technical updates & work on tests
More than 2000 journals added
Around 150 new normalized organizations
99.99 million (!) articles
32.4 million Open Access articles
Fix issue where ghost highlighting would appear on truncated abstracts
Some fixes to automatically clean up common query syntax errorsnull
Integration with PlumX metrics instead of Altmetric. Citation counts are present on many results
Better behavior – offering suggestions, attempt to show results even when a query cannot be parsed (for example: due to unmatched parentheses or quotation marks.
Fixed incorrect counts for autocomplete suggestions in Journal and Organization filters (note that the counts in the facets are still approximate rather than exact)
Fixed problems adding filters on an empty request (i.e. when you click the search icon without entering a query)
Fixed problem which occurred when Access Type filter was added, followed by modifying or reloading the query
Allow filters to work even if its all lower case e.g. AUTHOR: or author:
The help icon (search box) function better on mobile
Browser page titled were incorrect for 1foldr Hubs
Entirely redesigned user interface
Better technical foundation for user interface: changes will be faster and easier
Improved functionality for phones and tablets
New selection interface allows selection across searches
Filters and query syntax are unified
Added article detail pages
31M Open Access records
96M records in total
Added 400,000 records in 2019
Added Publisher filter
Added Journal Country filter (beta)
Some improvements to relevance ranking on searches that match the authors field
Fixed bug where, if switching quickly between articles and patents or changing filters quickly, incorrect results might display
Improvements for monitoring (alarms)
Fix for autocomplete input containing spaces
Fixed autocomplete input containing accents
Partial fix for issue where organization autocomplete would not show all results
Addition of many PR journals from Korea and Japan
Normalization of 100M more addresses from raw records
Harvesting of new Korean and Chinese data
Addition of new fields to the dataset: publisher country, organization country, and publisher name
Added Patent feature – users can search through almost 15 million U.S. and European documented patents. This feature is a new implementation, and we are consistently monitoring any disparities in data or UI which may occur – several that we are aware of are currently being resolved. We encourage all users to report any bugs they encounter while using the patent search!
Fixed an issue regarding logging in/out on different web browser tabs.
Fixed an issue where users were experiencing bugs when they would leave or enter an IP-authenticated campus with their laptop.
Allow login/logout from within 1findr interface
Better integration with Zotero and other reference managers
Add PDF URLs (via 1science resolver) to data export
Separate author first & last names, when available, in data export
Highlight matching authors, and always show highlighted author (even if the article has more authors than can be displayed)
Use author first/last names, when available, in citations
Refactor configuration system
Improve logging of application errors and access denials
Changes to 1findr
Add new journals to the whitelist
New taxonomy implemented
Many journal fixes (merging, renaming)
Add and normalized Publisher names
Add and normalized Journal countries
Remove many journals published by dubious publishers
Remove duplicate articles from 1findr
Our previous deduplication algorithm had 3 steps, we’ve now added 8 more steps for a total of 11 steps to have better chances of removing duplicates
According to our 500 duplicate reference sample, duplicates have gone from 8% to less than 2% (we remove more than 5M duplicates)
Display selected keywords in Show More
Display error message when we’re having trouble parsing a query
Tweaks to relevance algorithm
Fix occasional misinterpretations of hyphens as NOT operators
User Interface & Search API
Add new Journal filter, which both shows the top journals for your request and allows you to quickly search for a journal
Support searches by ISSN
Change relevance/ranking algorithm for author & title fields
Allow up to 50 results per page
Extend highlighting of search terms to authors and journal names
Better activity logging, to be able to remove invalid links in our data more efficiently, and to improve data update
Non PR journals removal in 1journal
New journals in 1journal
Normalization of journal titles
Remove duplicates in journals
Fixed an issue where several publishers weren’t receiving visibility: Wiley, Sciencedirect, Degruyter, Taylor and Francis, etc.
Bug fix: do not add editor’s names to authors
Bug fix: clean crossers article titles
Rework relevance algorithm, incorporating citation data
Add endpoints to fetch references and citations
Rework query fuzziness logic so that if e.g. you paste a long title, not every single word needs to match, but all explicit field queries always need to match
More intelligent searching for hyphenated words (“e-mail” will match “email”)
Tweak stemming behaviour, and attempt to deal with some English spelling variations, so that a search for “externalizing behaviours” will match documents for ‘externalizing behavior’
Generated Advanced Search queries with fewer parentheses
Fix Google Analytics logging of user data
In metrics, log a unique ID for the search, and log the position of results where users click on PDFs
Fix appearance of some dropdowns in IE1
Add search unique ID so joining search and resolver data will be easy
Protect profiles admin web console with AWS’ firewall (WAF)
Searches should yield more articles with abstracts
Empty searches no longer result in an error message
Add more logging to metrics (so we can finely monitor what domain names return high error rates)
Cascading resolution: we now use all pdf links instead of the first one
Added 2 million organizations in articles
Support for free version
Support for free version
Support for free version (usage plans for anonymous users)
Added 1 million articles
Increase the number of abstracts by 50%
Increase journal coverage (+1000 journals in 1journal)
Decrease the number of duplicates articles in the data
Improve link freshness for ResearchGate
Add addresses (organizations)
Use Article Referential as a source of data, instead of raw article files
Add articles from OJS journals
Organizations (Beta) facet added
Feature: Organizations for an article displayed in Show More
Feature: Organization added as field in Advanced Search
Feature: Altmetric patent citation data displayed
Feature: Allow custom headers for repositories like http://malaria.1science.com
Fix: Big display bug in IE11
Fix: Data exports failed in IE11 and Edge
Fix: In Subject filter, error trying to deselect an only child
Fix: Errors in Access Type filter on OAfindr
Feature: Made organizations searchable
Fix: Issue with result sets on some custom repositories
Allow domain names whitelisting through profiles
1findr and 1foldr
The platform is now easier to maintain, more secure, and we have much more logs to trace access
Our user profile back office system is now fully deployed, enabling easy creation & configuration of 1foldr hub products (such as https://alzheimer.1science.com/home/ and https://ph-sp.1science.com/home/)
Fixed issue with data export of CSV, TSV, and RefWorks in Chrome and Safari
oaFindr now uses the new user interface (same as 1findr). It can be completely configured through out back office web interface.
Added UI for OAfindr (Open Access articles only), improved user-metrics tracking
Fixes to author search with wildcards and DOI search, improved user-metrics tracking
Add multiple options for citation styles in Export dialog
In facets, update counts based on other active facets (so that if you e.g. check a year, you’ll see the correct counts for access types for that year)
Upgrade Elasticsearch from version 2 to version 6
Significant simplification of structure of indexed data
Accent folding: searching for Éric also finds Eric, and vice versa
Bigram signals in relevance: when typing several words in the search box, how many of those words show up in sequence in the results is taken into account in relevance ranking
Two new 1foldr hub repositories
New version of the search API
Added NOT operator to Advanced Search
Fix Medline DOIs
300,000 more articles in 1Findr (from 88,219,583 to 88,507,885)
Added journals and articles – 1findr articles have reached 88 million!
Enhancement: Changed Year filter design to fix validation issues
Enhancement: Indicate active filters
Enhancement: Disable Reset Filters when no filters are active
Fix: Some data contained HTML, which was being rendered in the UI
Fix: When changing search page, scroll to the top
Accessibility: Added several skip links
Accessibility: Many small changes based on feedback from government
Enhancement: Pre-fill custom query when you go there from the query builder
Enhancement: Integrate new Govt of Canada header, use their language-switcher link
Enhancement: Implement library link resolvers
Enhancement: Several design changes to results
Enhancement: Add Altmetric data to results
Enhancement: Add an internal help page for the government
Fix: Clear Advanced Search when you edit the query elsewhere
Fix: Translations were being cached in-browser and thus not updating
Fix: Subject filter is translated in French
Fix: Broken Advanced Search queries
Fix: Opening citation window would unnecessarily scroll to the top
Fix: Scroll to the top after performing Advanced Search / changing sort setting
Accessibility: Options in dropdown lists are not read by screen readers
Accessibility: Dropdown select lists do not work with keyboard when screen reader active
Accessibility: When screed reader active, the language filter selection fails
Accessibility: Navigation inside subject tree broken when screen reader active
Fix: In the Organization filter, don’t show organizations with 0 results
Fix: Subject filter was broken when you selected two subjects at different levels of the hierarchy
Fix: log usage metrics properly
Setup resource monitoring
https://search.1science.com/api/v2/1findr allows to search through the entire dataset of 1findr.
Date : 19/12/2017
New version of data added
Enhancement: in Subject filter, revamp selection behaviour, introduce partially-checked state
Enhancement: translate Organization filter values
Enhancement: Open All Selected tries to be more intuitive when selecting some non-OA results
Enhancement: move buttons on Advanced Search
Fix: In IE11, some search result links were not visible
Fix: inability to scroll to the top of page after exporting data or selecting results
Fix: inability to click Author/Journal links to filter
Fix: IE11 word-wrapping in subject filter
Fix: Show More link would trigger an unwanted scroll up
Fix: Subject-filter counts cut off on Windows
Fix: Open All Selected button did nothing
Fix: after reload, could not click buttons on Advanced Search page
Fix: in Subject Filter, counts would not update on new query
Accessibility: make Advanced Search language tag removal keyboard accessible
Accessibility: keyboard access on language menu
Accessibility: Instructions on how to complete language field unclear
Accessibility: Instructions for Subject field not clear
Accessibility: Show More link not keyboard-accessible
Accessibility: Radio buttons require fieldset
Accessibility: Add various ARIA labels
Journal search now works with journals names that contain special keywords like “and”, or “of” (https://1findr.1science.com/search?query=journal%3A%28Advances%20in%20Materials%20Science%20and%20Engineering%29
Make theming possible
Fix a bug where sometimes “No results found” would flash briefly when there were results
Update to oaFindr index, we’ve now reached 29M articles (previous release was at 28.8M)!
Fixed a bug related to publication year issue in Crossref
OA colors have been fixed (Gold: 14M, Green 9M, Unknown 7M)
Crossref and Medline are now up to date
Support to filter out social media networks
46 new journals
Article database: we are now indexing 28.8M articles (previously was 28.3M)
PDF URL ordering: Improved link prioritizing
Organizations / institutions: orgs have been added to our index, for Canadian scientific production only.
The newest version of our index now offers nearly 25.5 million scholarly articles. This small oaFindr release provides mostly bug fixes, with new features and a new look for oaFindr in development.
Fixed a bug which truncated Article Title on Some Results
Fixed a bug where endnote Imports Only First Article
MeSH Terms: Articles from MEDLINE now include their Medical Subject Headings (MeSH) as keywords, to improve their discovery.
Search by Keywords: It is now possible to search by the keywords associated to an article, directly in the search box.
Advanced Search by Title, Abstract and Keywords: A new option has been added to the query builder in the Advanced Search, allowing you to search through a combination of an article’s title, abstract and keywords. This combination of fields provides optimal results for a query.
DOI Search: A bug which our search backend stopped recognizing Digital Object Identifiers in a query has been fixed.
Filter by Journal Name with Syntax Characters: We discovered that some journal names contain characters, such as slashes and ampersands, that our search backend would interpret as syntax characters for a complex search. Filtering your search results by such a journal name would result in an error message. This h
Filter by Open Access Type: 80% of the articles are now classified in our index based on the type of Open Access under which they are offered: Gold or Green. Results can be filtered by either or both types as well as include it as a search criteria in the Advanced Search. A helpful icon next to each result’s title will also indicate which type of Open Access that article is available under.
Better Handling of Undetermined Languages: We’ve improved how we classify article languages, which means a more accurate count for those articles where we can’t yet identify the language and better filtering in our aggregation panel.
More Browsers: We always try to provide the best user experience in oaFindr and that has required a modern web browser. However, we were rather conservative in which browsers had access to the application. We’ve now broadened our support and welcome users of Opera, Vivaldi and other modern browsers to oaFindr.
Copy Link Without Flash: The action button associated with each search result always offered a way to copy the article’s link to your clipboard, but it would use Flash in an attempt to do so. Our new solution does not need it anymore, which makes oaFindr Flash-free!
Tooltips: Some helpful tooltips on some elements of the search result page were missing. We’ve added them and hope they help.
Number of Search Results per Page: It was possible to alter the number of search results per page directly in the URL, but setting too high would cause the application to freeze, and that’s bad. We’ll now prevent a number of results that would degrade performance too much.
Exact Search: We always try to give you the best results for your query, but sometimes you know exactly what you want. Surrounding one or more search terms with double quotes (“”) will now return exact matches for those terms.
Faster Search: We’ve revamped our search system in a big way, which means both faster search and more accurate results.
Improved Articles Deduplication: Finding duplicate articles is a waste of your time, so we’ve improved our deduplication process to filter more of them out. We now have over 18 millions articles in our index, with fewer duplicates than ever.
Abstract Preview: Abstracts are a great way to find out whether an article is pertinent to your needs. We now show you the first three lines of an article’s abstract right in the search results, to make it easier or you to weed them out.
New Languages Filter: Our Language Filter is out of beta! It will show you the top five languages for the results of your query, with an option to expand the list and see all the languages represented in your result set.
New Years Filter: We’ve also improved the way the Years Filter works. We now show you the five latest years that have results in your search, with an option to expand the list and see all the years with articles in your result set.
New Message When Search Has No Result: When a search returned no results, we used to display a full page informing you of the fact. We’ll now just display a message, rather than send you away, letting you tweak your query to get better results.
Easier Date Selection in Advanced Search: Selecting a date range in our Advanced Search page was cumbersome, so we’ve made it easier to pick a range, or to choose from some common, useful ranges: Current and Past Year, Last 5 Years and Last 10 years.
Selecting All Subjects: When you selected all the subjects one by one, an empty subject criteria would be added to your query, and an error message would appear. We’ve taken care of that. Back Button Issues: We discovered one case where using the back button would not return you to your previous search, but reissue your current one. That’s now fixed.
Third Level of Subject Filters
With the objective of providing our users more precise results to their queries, we have implemented a 3rd level of subject headings. Due to the new level of article classification, we had to completely overhaul the way users select subjects to fit within the design of the site.
Filtering by article subject is now a process of selecting headings within a classification tree. Clicking on a check box next to a subject heading selects all articles under that heading. If the user wants to be more precise, they can click on the arrow next to the number of results to reach the next level of classification where they can choose the fields they want to search.
Once an arrow is clicked, a layer of related headings is shown, allowing the user to further narrow down their selection.
These updates on the search page also carry over onto the Advanced Search page where the user can navigate the full tree to select their fields in a more visible manner.
As this feature is rather dynamic we strongly encourage that you check out the new filtering systems on the site to get a better idea of how to filter by subject headings.
Index increased to over 12.5 Million articles
With a few adjustments made to our pipeline we were able to acquire over 2 million new articles which will soon be incorporated into our live index.
OaFindr is now available in Spanish
All feature from the search, the Help Center, and the attached chat have been translated to Spanish, making it the 3rd language available to our users.
Switching languages is easily done on any page of the product by clicking on the language the page is currently being displayed in. On the landing page there are two different arrangements of language selection. The first is in the top right corner the second is in the bottom left hand corner:
When clicking on the language in the top right hand corner the user is shown a pop-up where they can choose which language they would like to use the website in.
Changing the language of the website does not influence results returned after a search. However, when the language of the website is used in tandem with the language filtering feature introduced in the release notes for version 1.34 the user can have a full experience of using the website completely in Spanish.
Our help center has also been translated for easier use by our Spanish speaking users.
oaFindr now supported on Linux OS
Previously oaFindr could not be used on Linux based computers however this has been fixed with the most recent update.
Release 34 introduces several major language features to oaFindr
Language Filtering in Facets
Due to popular demand, 1science has launched the Language Filtering feature which provides users the ability to refine and increase the relevancy of their results based on the language of the article. This will make it easier for users to find the resources they need.
When a user performs a search they can choose to filter their results with the language option by
1. Clicking on the ‘Languages’ facet in the left hand side bar
2. Selecting preferred language
The initial view of the language facet will list the top 2 language results and display the number of articles found where the language is Undetermined.
(We are a step ahead: our team is already working on further identifying the language of articles that are currently undetermined).
Once the user clicks on ‘+ All languages’ a full list of retrieved articles in the remaining languages will be shown in alphabetical order.
Users will now be able to select multiple languages so that they are only shown results in those particular languages.
Language Filtering in Advanced Search
With the addition of language filtering on the Search Engine Result Page (SERP), users can now also search by language on the Advanced Search page. Here, users can pre-select what languages they want results to appear in. 6 Enhancements
The left hand side of the results page has seen some minor changes. Along with the addition of Language to the filtering options, Years has been moved to the top, as it is a more useful filtering option than Subjects.
Release 30 introduces several new features to oaFindr to increase accessibility and the user interface.
Cite / reference Copy-Paste
Save time citing references with the new cite / reference copy paste feature. The reference style is displayed in a Chicago Manual of Style 16th edition. To copy and paste references, follow these steps
1. Select the … symbol
2. A drop down menu will appear, Select Cite
3. A window will appear with the citation in the Chicago Manual of Style
4. Simply copy the citation and paste with the rest of your references
Data Change Request
We try to be perfect and get everything right but sometimes some mistakes can slip through. If you stumble on erroneous metadata or information that seems bizarre to you, you are now able to report these issues. Simply click on “Report” and fill out our form.
We have enhanced the look and feel of the platform with home, headers, footers, facets, revision buttons and introduced a minimalistic design with emphasis on ease of use in order to improve the functionality of the platform.
Sorting facets are now expanded by default
We added the option to sort by “Oldest article on top”
To discover this new enhancement, find it below the sorting facets on the left-hand side.
The Advanced Search has now been optimized to support combined searches and repaired language operators Improvements include:
Boolean expression in the search fields
Topics and dates are isolated from the construction of the request
Always visible Search button
Improved functionality on mobile
Example queries by type of fields
Bring out the loader! The orange bar on the top of your page is now bolder and is now more visible so you can clearly see it processing your search query.
Check boxes to select an item
Check boxes to select an item are now always visible to allow an easier and streamlined selection.
Fixes Export citation content
The export quality of RIS, Zotero, RefWorks, Bibtex, CSV and TSV formats has been improved
Opening facets in Chrome OS X
Fixed facets behaviors in Chrome OS X
Improved highlighting and relevancy in the SERP
When searching for a phase (group of words) your search term is now bolded in the abstract to indicate the relevancy of the article
Global relevancy has also been improved
This post is also available in: French