Embedded Files
In recent years, more and more library vendors have incorporated AI-based technologies in their products and services. Furthermore, there is a trend of more library vendors are exploring and implementing these emerging technologies. We further notice that more vendors are strategically making use of AI-related buzzwords, such as AI, machine learning, deep learning, natural language processing, and data analytics, in their marketing practice. Library professionals are more interested in learning and trying these AI-infused products. With concrete examples and use cases, we hope to demystify AI-associated technologies. In this section, we highlight a sample of these vendors and their products. The list is far from complete, and our selection is by no means representative and scientific.
Data Mining
The St. Paul, MN based company offers accurate and streamlined content connection between library patrons and library resources. Third Iron employs AI in the form of an Expert System in building direct link to articles. This linking service takes into consideration:
Knowledge about the articles (is this an article in press? Does it have complete metadata?)
The access status of the article (OA for the journal? Just this article? Embargoed?)
The destination for the article (an aggregator? the publisher? Which platform?)
The holdings available to the library from multiple sources
The authentication routes for the user
The user’s IP address (if IP authentication is being used)
ILL/Document delivery resources available
Digital Preservation
Based in Madrid, Spain, the company assists organizations in preserving their content by applying AI techniques such as automation of ingesting processes, content analysis of digital objects, automated classification of digital files, optimizing trustworthy data storage, and metadata generation.
Discovery Tools
Owned by the Holtzbrinck Publishing Group in Stuttgart, Germany, Digital Science was launched in 2010 with a goal to make research lifecycle more efficient, effective, and open for the scientific and research communities. The company’s portfolio consists of a range of tools, services, and brands, including Altmetric, Dimensions, Anywhere Access, Figshare, ReadCube, Symplectic, IFI Claims, GRID, Overleaf, Labguru, BioRAFT, TetraScience and Transcriptic. It’s flagship product, Dimensions, is said to be the world’s largest linked research information dataset and the most comprehensive research grants database. Dimensions links grants to millions of resulting publications, clinical trials and patents. It covers millions of research publications connected by more than 1.3 billion citations, supporting grants, datasets, clinical trials, patents and policy documents. Using machine learning and cloud computing technology, Dimensions integrates, aggregates, and transforms data from myriad sources to create a consistent model. An introductory video is available at: An introduction to Dimensions. Dimensions is available for a test drive at: Search Dimensions.
As a subsidiary of the EBSCO Information Services, DynaMed “publishes clinically organized topics which are continuously updated as new information indicates a need for change in practice or improved understanding of the body of knowledge.” Partnered with IBM’s Micromedex with Watson, DynaMed offers curated drug and disease content into a single source for evidence-based insights for informed clinical decision. Its content ranges from comprehensive reviews of diseases, conditions, and abnormal findings to highly focused topics on evaluation, differential diagnosis, treatment options, and management. Through its website, healthcare providers and pharmacists will type clinical questions, using natural language queries, into AI-powered Watson Assistant and get most relevant answers and most current clinical evidence.
Based in Ottawa, Canada, the company offers DistillerSR, an AI-powered classification manager and screening that automatically classify and deduplicate references and literature reviews based on customer’s specifications. DistillerSR competes with other systematic literature review products on the market, including Pico Portal, Covidence, EPPI-Reviewer, and SWIFT Active Screener. Several of them are also powered by AI and machine learning technologies.
As a library software company based in Jerusalem, Israel, Ex Libris offers cloud-based solutions to help libraries organize, management, and expose their print, electronic, and digital content. The company has adopted AI technology in the following three products:
Dara (Data Analysis Recommendation Assistant)
Dara provides recommendations to improve Alma user’s workflows and better optimize their usage of Alma. It can suggest features and validates custom configurations. DARA generates up to 30 recommendations of each type, driven by smart decision-support engine using machine learning technology and smart algorithms.
Through natural language processing, Primo Voice Search allows users to speak their search terms using their device’s microphone
Central Discovery Index (CDI)
CDI is Primo’s comprehensive content index that enriches patron’s discovery experience through smart discovery services driven by artificial intelligence.
Based in Modena, Italy, Expert.ai is a publicly traded company (EXAI:IM). It leverages machine learning and knowledge-based AI techniques to help organizations to manage and make better sense of their information assets. The semantic analysis products and services it offers consists of semantic and natural language text searches, text analytics, development and management of taxonomies and ontologies, categorization, data and metadata extraction, and natural language processing. EBSCO have employed Expert.ai’s technology in enriching its metadata and semantic tagging and enhancing patron’s information search and discovery. SAGE Publishing deploys Expert.ai’s Cogito cognitive technology to power its content recommendation engine.
Based in Irvington, NY, InfoDesk is a data service provider that employs proprietary natural language processing in acquiring, aggregating, and integrating content from a large group of content sources and connectors through entity recognition, concept disambiguation, and semantic indexing processes. Enriched information is delivered to customers based on their specifications and needs.
With headquarters in Norway, Finland, Bulgaria, and Ukraine, Iris.ai aims to become a comprehensive search platform meeting organization’s research needs. Its AI search engine uses natural language process and machine learning to review massive and diverse literature, extract key data, identify the most precise pieces of knowledge, generate summaries, and systematize data (Breeding, 2022). The company has partnered with University of Helsinki Library to enhance its research discovery operations. In another project, funded by the Norwegian Research Council, Iris.ai is collaborating with Open University, Oxford University, Trinity College, Duplin and University College, London in developing AI Chemist, a discovery platform of chemistry literature (Notay, 2022).
This New York-based service grew out of the backdrop of COVID-19 pandemic in 2020 to address the growing need for an effective tool to speed up the process of synthesizing scientific literature in biomedical research. Based on machine learning, natural language processing algorithms, PICO Portal is a web-based systematic review management tool that transforms research findings into relevant and high-quality evidence. Its major competitors include DistillerSR, Covidence, EPPI-Reviewer, and SWIFT Active Screener.
Based in Brooklyn, this startup helps researchers better discover and understand research articles through Smart Citations, citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. Smart Citations allow users to see how a publication has been cited by providing the context of the citation and a classification describing whether it provides supporting or contrasting evidence for the cited claim. Scite relies on its proprietary machine learning model trained on over 40,000 citation statements. Scite has received funding from the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
Based in Palo Alto, CA, Yewno was founded in 2015 with a goal to help “the world to uncover the undiscovered through its dynamic Knowledge Graph and AI based inference engine, which introduces an entirely new approach to knowledge extraction offers” (Breeding, 2022). As a semantic-analysis engine, Yewno’s mission is extracting augmented intelligence and knowledge from a vast quantity of unstructured and structured data by applying computational linguistics, network theory, machine learning, and methods from AI. At the core of Yewno’s proprietary technology is a framework that extracts, processes, links, and represents atomic units of knowledge from heterogeneous data sources. Its deep learning network continuously “reads” high-quality sources projecting concepts into a multidimensional “Conceptual Space” where similarity measures along different dimensions are used to group together related concepts. According to Yewno, such “Conceptual Space” is built on the cognitive theory, allows for both geometrical, statistical and topological operations, and permits to aggregate basic concepts into more complex representations (https://www.yewno.com/about).
Yewno’s suite of solutions includes:
Yewno Discover: “Yewno’s Discover platform harnesses hundreds of millions of semantic connections and conceptual links from millions of scholarly articles, books, and databases across all academic fields. This empowers students and researchers to navigate intuitively across concepts, relationships, and disciplines, learning from resources that might have otherwise been overlooked. This not only enhances understanding and creates more impactful work, but also saves time while ensuring comprehensive and credible coverage.” (https://www.yewno.com/education)
Yewno Unearth: “Yewno Unearth provides a semantic snapshot and insights into cross-disciplinary trends, enabling students to better understand and leverage their collections. Yewno’s Unearth product is designed to facilitate understanding, categorization, and cataloging.” (https://www.yewno.com/education)
Yewno | Edge: “AI-Driven Investment Research Platform. Yewno embraces the overconnected world, finding relationships in big data that help you make better decisions.” (https://edge.yewno.com/)
Yewno Discover has been adopted as an alternative to the standard library catalog or discovery service by academic and public libraries. Below are highlights of those early adopters:
University of Edinburgh Library: It is the first university in Scotland to adopt Yewno Discover platform in 2022 (Breeding, 2022). According to library’s website, “Yewno Discover helps researchers, students, and educators further explore knowledge across interdisciplinary fields, sparking new ideas and finding unexpected connections along the way. Its visual search interface illustrates the links between topics and concepts across all academic fields and helps to identify the literature that joins them from millions of scholarly articles, books, and databases.” (https://www.ed.ac.uk/information-services/library-museum-gallery/discovery/yewno-discover). Edinburgh’s Yenwno Discover, which requires login, complements the library’s Primo Discovery service.
University of California, Riverside (UCR): Introduced in April 2022, UCR’s Yewno Discovery is fully integrated with its Primo Discovery service, WorldCat Discovery, and other resources (UC Riverside Library, 2022). It is limited to its own students, faculty, and researchers. To help users using this new too, the library further offers a Yewno Discovery Guide.
Stanford University: An early adoption of Yewno, Standard has been using Yewno Discovery for more than five years. On Stanford Library’s website, Yewno is described as “a discovery tool that provides a graphical display of the interrelationships between concepts.” It “uses computational semantics, graph theory, and machine learning to extract concepts from scholarly publications … and displays search results in a graphical interface that displays the interrelationships between those concepts.” Following links in Yewno search display, users will “find the underlying publications from which concepts and relationships are extracted.” Yewnow is further integrates with library subscribed resources to allow users to access to the full text of licensed resource directly (Stanford University Libraries, 2022).
Santa Clara County Library: Available since 2019 at the Library, Yewno Discover is described as an AI-powered discovery and learning tool for cross-disciplinary search. It allows users to browse results through a a visual map, called a Knowledge Map, that depicts interrelationships among concepts (Santa Clara County Library, 2022).
Yewno offers abundant training and technical materials at its YouTube channels. The following videos provide a good overview of its products:
Yewno Discover Basics Part 2: Saving and Exporting Search Results
Yewno Discover Basics Part 3: Exploring Layers and Generating New Relationships
Other AI-powered Discovery Tools
Below is a list of AI-based search and literature mapping tools we have identified. Most of them are start-up companies with small customer base. Most of them offer free account that does not include full-range of service. Due to time constraints, we are not able to conduct a thorough research, testing, and comparison of these services.
Citation Gecko: “Gecko is here to help you find the most relevant papers to your research and give you a more complete sense of the research landscape”
Connected Papers: “Explore connected papers in a visual graph”
Consensus: “Better science-based answer, faster”
Covidence: “Better systematic review management”
Feedly: “Track insights across the web without having to read everything”
Inciteful: “Tools to help you accelerate your research”
Litmaps: “Visualize, expand, and communicate your research expertise”
Open Knowledge Map: “A visual interface to the world’s scientific knowledge”
Research Rabbit: “We’re rethinking everything: literature search, alerts, and more”
ResearchGate: “Discover scientific knowledge and stay connected to the world of science”
Semantic Scholar: “A free, AI-powered research tool for scientific literature”
Expert System
Found in 1989 and based in Encinitas, CA, Soutron has been a vendor of integrated library system for special libraries. Most of its customers are corporations, law firms, or healthcare organizationa. The Soutron ILS is the company's flagship product. Additional complementary products include a Soutron Archive, the Soutron Digital Media Archive, software for records management, and the Soutron Discovery platform (https://www.soutron.com/products-usa/)
Soutron’s Ask Soutron, launched in 2020, is an enquiry management plugin powered by AI and cognitive intelligence technologies. It provides automated assistance to Knowledge Service, an enhanced Artificial Intelligent driven query tool to support online enquires. The inquiry response service incorporates natural language search with fast performance that executes against internal and external data sources, websites, and other resource sets.
Learning Analytics
Based in London, UK, Bodyswaps offers library customers training solutions on soft skills in the areas of job interview, public speaking, workplace communications, teamwork, customer service, inclusive leadership, and diversity. Leveraging virtual and AI, Bodyswaps training simulates realistic scenarios and allows students to learn in safe and realistic environments. Bodyswaps features its “AI-powered behavioral analytics engine [that] provides learners with personalized assessments,” which are then feeds back into institution’s learning management system for real-time soft skills competency tracking. Furthermore, the virtual reality platform allows students to reflect in the context of practice through an immersive virtual environment and to apply to their own situation. Bodyswaps has formed a partnership with SAGE Publishing to launch a virtual reality training resource for health and social care learners (Breeding, 2021).
References
Andrew, W. (2021, February 3). Quick answers: How to define artificial intelligence. Gartner. https://www.gartner.com/document/3996413
Breeding, M. (2021, December 2). Sage Publishing and Bodyswaps Launch Virtual Reality Resource for Health and Social Care Training. Library Technology News Service. Retrieved from https://librarytechnology.org/pr/26874
Breeding, M. (2022, January 24). University of Edinburgh adopts Yewno discover to offer AI-powered research through their libraries. Library Technology News Service. Retrieved from https://librarytechnology.org/pr/27021
Notay, B. (2022, February 21). Iris.ai and CORE cooperate to build AI Chemist. Jisc Research. https://research.jiscinvolve.org/wp/2022/02/21/iris-ai-and-core-cooperate-to-build-ai-chemist/
Pelletier, K., McCormack, M., Reeves, J., Robert, J., & Arbino, N. 2022 EDUCAUSE Horizon Report, Teaching and Learning Edition. Boulder, CO: EDUCAUSE. https://library.educause.edu/resources/2022/4/2022-educause-horizon-report-teaching-and-learning-edition
Santa Clara County Library. (2022). Yewno Discover. https://sccld.org/yewno-discover/#top
Silipigni Connaway, L., Cyr, C., & Gallagher, P. (2020). Global perspectives on discovery and fulfillment: Findings from the 2020 OCLC Global Council survey. OCLC. https://www.oclc.org/content/dam/research/publications/2020/oclcresearch-global-council-discovery-fulfillment-survey.pdf
Stanford University Libraries. (2022). Yewno. https://library.stanford.edu/search-services/yewno
UC Riverside Library. (2022 April 29). UCR Library implements Yewno Discover for AI-powered research. https://library.ucr.edu/about/news/ucr-library-implements-yewno-discover-for-ai-powered-research
Yanckello, R. (2021July 13). Hype cycle for higher education, 2021. Gartner. https://www.gartner.com/interactive/hc/4003518?ref=solrAll&refval=330117274
Page updated
Report abuse