Library Help: Office For Open Research

Understanding research classification 

Research outputs are classified in various ways across different platforms, helping to organise scholarly works and to identify patterns in research activity. These classification systems range from broad subject categories to highly specific research topics. 

Major classification systems 

Web of Science categories 

• Hierarchical structure with 254 subject categories 

• Groups publications based on journal classifications 

• Allows analysis at both broad and specific subject levels 

• Particularly useful for understanding traditional disciplinary boundaries
   

OpenAlex Research Topics 

• Uses a machine learning approach to classify research 

• Provides multiple levels of classification 

• Works with concepts rather than traditional subject categories 

• Offers more granular and flexible classification than journal-based systems 
   

Australian and New Zealand Standard Research Classification (ANZSRC) Fields of Research 

• Structured hierarchical system developed jointly by the Australian Bureau of Statistics and Statistics New Zealand 

• Two, four, and six-digit codes representing different levels of granularity 

• Widely used for research assessment and funding allocation 

• Particularly valuable for international benchmarking 

• University of Manchester publications can be viewed under these classifications through both Altmetric Explorer and SciVal platforms 

SciVal Topics and Topic Clusters 

SciVal offers a sophisticated approach to research classification through its system of Topics and Topic Clusters. This dynamic classification method can provide valuable insights into research themes and trends. 

Understanding SciVal's classification system 

Topics (About 94,000) 

• Smallest units of analysis 

• Generated using citation patterns and natural language processing 

• Updated regularly to reflect emerging research areas 

• Each topic contains a collection of papers with similar research interests 

Topic Clusters (About 1,500) 

• Groups of related Topics 

• Provide a broader view of research areas 

• Help identify larger research themes and patterns 

• Useful for strategic planning and collaboration identification 

Elsevier guidance on Topics and Topic clusters: this guide provides more detailed information relating to the methodologies used in the formation of topics and topic clusters. 

Useful features 

• Topic prominence: this provides an indicator of momentum with a particular research topic (using a combination of citation and views-based metrics to help identify ‘hot’ research areas). 

• Keyphrase analysis: Uses Elsevier Fingerprint technology to extract distinct keyphrases within a given entity (including topics and topic clusters). These can be displayed in tabular or visual format.

Practical applications 

These classification and analysis tools can be used to: 

• Inform research strategy: by identifying emerging research areas, tracking the evolution of research fields and spotting gaps in current research coverage. 

• Find collaboration opportunities: by identifying leading institutions in specific topics, spotting cross-disciplinary opportunities and mapping collaboration networks 

The Research Indicators team can help you to navigate different classification systems and create custom reports and visualisations. These can help to identify strategic opportunities and identify potential collaborators. 

We have made successful use of SciVal Topics and Topic Clusters to create a network of research publications in the fields of Artificial Intelligence and health equity. These allow for easy identification of researchers from various schools and faculties who share similar research interests. 

Please Contact us if you wish to learn more about the various options that are available.