ResearchPulse automates the Research Impact Assessment Framework (RIAF) by combining artificial intelligence with systematic data analysis of academic publications, web content, and institutional data. We help researchers and funding bodies generate comprehensive use-case studies and research impact assessments.
🎧 Listen to the ResearchPulse DeepDive AI Podcast - How AI is transforming research impact assessment.
Our intelligent system combines four powerful components:
-
Smart Publication Analysis
- Multi-source academic database integration (ArXiv, PubMed Central, Semantic Scholar)
- Intelligent publication filtering and citation-based ranking
- Automated PDF processing and metadata extraction
-
Web Intelligence
- Real-time web content analysis
- Multi-format content processing
- Automated relevance assessment
-
Data Catalogue Integration
- Automated processing of health and medical data catalogues
- Evidence-based impact metrics
- Intelligent data relevance ranking
-
Local Document Processing
- Seamless integration of your research documents
- Multi-format support (PDF, DOCX)
- Automated metadata extraction
- Time-Efficient: Transform weeks of manual analysis into comprehensive use-case studies within minutes
- Evidence-Based: Automatically gather and validate evidence from multiple sources
- Customisable: Adapt assessment criteria to your specific research programme
- Comprehensive: From problem definition to future applications
Whether you're:
- Preparing funding applications
- Documenting research impact
- Generating use-case studies
- Assessing research outcomes
ResearchPulse provides automated, evidence-based analysis while maintaining academic rigour and relevance.
What is ResearchPulse?
ResearchPulse is an AI-powered tool designed to automate the generation of research impact assessment use case studies, specifically for health and medical research. It applies the Research Impact Assessment Framework (RIAF) to evaluate research environments and their influence within a specific research ecosystem. The tool generates detailed use cases documenting research impact, aiding funding organizations in assessing specific research programs and their outcomes.
How does ResearchPulse work?
ResearchPulse employs a Retrieval-Augmented Generation (RAG) pipeline. It retrieves data from various sources like academic databases (PubMed, arXiv, Semantic Scholar), web content via Bing Search API, local documents, and structured health data catalogs. It then processes this information through a context engine, vector index, query engine, and review system. Finally, it generates a comprehensive research impact assessment report with references.
What are the key features and benefits of using ResearchPulse?
ResearchPulse offers a range of features including automated generation of use case studies, data processing from multiple sources, customisable research indicators, multi-format report generation, and a user-friendly web interface. The benefits include a systematic and structured approach to research impact assessment, saving time and resources, and providing evidence-based analysis to support funding decisions.
How can I get started with ResearchPulse?
You can access ResearchPulse through a hosted web interface or by installing it locally. Currently the web app requires a university network connection and login. Local installation requires a Python environment, API keys, and university network access. Detailed instructions for both methods are available in the documentation.
What kind of output does ResearchPulse generate?
ResearchPulse generates a downloadable zip file containing multiple outputs designed to ensure complete documentation and transparency. At its core, you'll receive a detailed research impact report available in both Word and Markdown formats, accompanied by context analysis documentation that details the research environment. The system also preserves individual answers for each assessment question, including both draft and final versions, allowing you to track the development of the assessment. To support transparency and verification, the package includes all source documents used in the analysis, along with a structured table of any documents that couldn't be automatically accessed (tracked in missing_documents.xlsx). For process validation and reproducibility, the system provides detailed logs tracking all system operations and AI process steps. All documents are exported in accessible formats, and a record of input settings ensures the assessment process can be reviewed or repeated if needed.
Can I customise the research impact assessment criteria?
Yes, ResearchPulse offers multiple customisation options to ensure your impact assessment aligns with your specific research context. Users can define custom time periods for both research activity and impact measurement, allowing for precise temporal analysis. The system's flexible framework enables you to select and prioritise relevant research indicators, whether you're focusing on academic impact, societal benefits, or economic outcomes. You can tailor the assessment criteria to align with your research goals and funding requirements, including specific institutional or grant body frameworks. This customisation extends to language style preferences, helping you generate reports that match your intended audience's expectations. The system also allows you to incorporate additional context through supplementary documentation and custom indicators, ensuring your unique research narrative is accurately captured.
How does ResearchPulse ensure responsible AI practices and ethical use?
ResearchPulse implements several key measures to ensure responsible AI use and ethical data handling. At its core, the system uses Retrieval-Augmented Generation (RAG) to ground all outputs in verifiable sources, ensuring that generated content is consistently backed by evidence. This approach is reinforced by a multi-stage review process that validates the accuracy of all assessments. To address potential biases, ResearchPulse integrates multiple diverse data sources and employs intelligent filtering mechanisms. The system maintains a human-in-the-loop review process, allowing researchers to validate and customise assessment criteria based on their specific field and requirements. Transparency and accountability are maintained through extensive process documentation and logging. Data sources are clearly attributed, ensuring users can trace any claim to its origin. Regarding security and user control, ResearchPulse operates within strict network security restrictions and processes only publicly available research data. Users maintain control over input parameters and can observe how outputs are generated, ensuring transparency throughout the assessment process. We continuously evaluate and improve these measures through user feedback and systematic monitoring, ensuring that responsible AI principles remain central to ResearchPulse's development and operation.What are the future development plans?
Future development plans include enhanced security features like SSO integration and private cloud deployment options, integration of international health data catalogs and media impact scoring, and expansion to other research domains beyond health and medical research.
How can I contribute to development or report issues?
All development activities and issue reporting are managed through a GitHub repository at the University of Sydney. Users with access can contribute to the project and report issues or feature requests through the repository's issue tracker.How can I get involved or contact the team?
We're actively seeking researchers to test and provide feedback on ResearchPulse. If you're interested in becoming a tester or have questions about the platform, please contact our friendly RIAF program manager Kirsten Jackson at [email protected]. We particularly welcome feedback from researchers in health and medical fields who regularly prepare impact assessments or funding applications. For technical questions or integration options into your system, please contact our main software author Sebastian Haan at [email protected].Main contributors at the Sydney Informatics Hub:
- Sebastian Haan
- Nathanial Butterworth
- Gordon McDonald
This project has been developed in collaboration with the Faculty of Medicine and Health at the University of Sydney and NSW Health, in particular:
- Kirsten Jackson
- Janine Richards
- Mona Shamshiri
Acknowledgments are an important way for us to demonstrate the value we bring to your research. Your research outcomes are vital for ongoing funding of the Sydney Informatics Hub.
If you make use of this software for your research project, please include the following acknowledgment:
“This research was supported by the Sydney Informatics Hub, a Core Research Facility of the University of Sydney."
Transform your research impact documentation with ResearchPulse – where AI meets academic excellence.