SPi Global

Leveraging AI/ML To Increase Content Discoverability

Posted by Sudhakaran Jampala Prasad


Posted on 19 February 2021



Content discovery, Content Creation


The Need For Efficient Content Discovery

For researchers, the ability to access the right content quickly and on-demand is essential for the success of their projects. Not only does the right content help them in their own quest to advance knowledge, but can have real-time impact on the world too. For example, researchers who are studying COVID-19 need to be able to locate the right and reliable knowledge or information in the quickest time possible.

With the amount of information that is available on the web, it is very likely that researchers may end up with content not relevant to them. Such instances where researchers are not able to discover a publisher’s content efficiently, it is a frustrating and futile experience for both stakeholders, irrespective of how good the published content is. Also, the absence of a single or unified discovery service that indexes all publications have further traditionally complicated content discovery for researchers.

In essence, the need for more efficient content discovery tools is now more acute.

How Latest Technologies Can Help

It is now evident that technologies such as artificial intelligence (AI) and NLP will allow researchers to access relevant content easily. Typically, the content discovery process is layered and complex. Content is generally spread across a publisher’s web page, across aggregators (e.g., ProQuest) and specialized subject matter repositories (e.g., PubMedCentral). Thus, the infusion of

AI and NLP will increasingly help researchers to search content in accordance with their preferences. Search can now happen according to specific publisher or author preferences through tools like automated intelligent recommendation functions, granular search functions, well-indexed metadata, etc.

This is easily conceivable since advanced computer vision technologies and sophisticated natural language processing can support researchers with the type of content that they seek with better reliability.

However, smart content discovery will depend extensively on well-crafted metadata so that researchers can scan expansive databases and, thus, subscribe to content that they find useful

Mobile Compatibility Is Indispensable

In keeping with the general trend, researchers too are expected to use smart phones more often to access links to research they want. This may cause more traffic to move away from conventional channels such as library digital catalogs due to both ease of accessibility and a more personalized user experience[1].

Influencing Is Taking Off

Recently, the industry is also using metrics such as altmetric tags to enable researchers to understand the influence that particular research content enjoyed among other readers. Analytics based on altmetric tags can provide a more integrated approach to understanding the online activity surrounding research content, which can help researchers to understand and select content they want.

Integrated Technology Approach- A Way Forward

The publishing industry will increasingly seek more advanced and integrated tools in concept extraction, NLP, complex machine learning, etc., for better categorization and delivery of on-demand content. Researchers can benefit from being able to link disparate pieces of literature from unrelated sources, thus paving the way for more significant interdisciplinary research

———————–

[1]https://content.iospress.com/articles/information-services-and-use/isu800

Contact Us

Any questions? Get in touch with us

0Shares
0