Your Privacy

This site uses cookies to enhance your browsing experience and deliver personalized content. By continuing to use this site, you consent to our use of cookies.
COOKIE POLICY

Skip to main content
Data & Analytics

Peer Recommendation Engine Designed for American Geophysical Union

Peer Recommendation Engine Designed for American Geophysical Union

By training a machine learning algorithm on the abstracts of previously published journals and live presentations, UDig designed a recommendation system to automate the process of matching peer reviewers for the American Geophysical Union (AGU).

How We Went from Ideas to Impact

  • By using the methodology developed by UDig, AGU ensures an equitable distribution of Peer Reviewers with representation across many demographics.

The Idea

Scientists from around the world submit articles to be published by the American Geophysical Union (AGU). Each of these articles must first survive a peer review, but the process to select individuals to review submitted content relied heavily on a human component to find appropriate authors. As a result, there was a narrowing of the scientists and authors most often selected to provide peer reviews which led to an overrepresentation of certain socioeconomic classifiers.

The Impact

Using the abstracts from previously published Journals and live presentations table, UDig designed an NLP-backed recommendation system. The NLP portion consisted of a term frequency-inverse document frequency (TF-IDF) model and a Doc2Vec model. TF-IDF is a measure used for information retrieval. Its intention is to reflect term relevance within a particular document. The idea behind TF-IDF is to assign importance when a particular word occurs multiple times within a document as it would appear that this word is meaningful within that document. At the same time, if the word occurs frequently in the target document as well as all other documents in the corpus, it will be assigned less weight as this may just be a frequently occurring word such as stopwords like “the” or “for”.

Doc2Vec’s purpose is to convert words or entire documents into numerical representations. It maintains order and semantic information of any arbitrarily sized text. In our doc2vec model, we used the abstract as the text corpus and the abstract ID to represent the articles associated authors. After text normalization, the modeling phase began. This phase consisted of hyperparameter tuning, training, and result evaluation. Both the doc2vec and the TF-IDF models compute similarity between the target document and the corpus. The abstract with the highest similarity score output by the models would represent our recommendation. Next, we randomly selected a list of 20 target abstracts for recommendations. We output 40 total recommendations: one from the TF-IDF and one from doc2vec for each target abstract.

AGU then had 21 different reviewers analyze the recommendations for relevance. The feedback was clear that TF-IDF outperformed the Doc2Vec model. By using the methodology developed by UDig, AGU ensures an equitable distribution of Peer Reviewers with representation across many demographics.

  • How We Did It
    Automated Taxonomy CreationRecommendation Engines
  • Tech Stack
    PythonAWSPostgres

Digging In

  • Artificial Intelligence

    RVA Data Enthusiasts | Accelerated Intelligence: AI Insights with UDig

    Join UDig to explore how AI is accelerating impact for clients. In this session, we will share technical insights and lessons learned across a variety of AI-driven projects. Through a round-robin format, you’ll hear from team leaders and consultants on key innovations, with time for Q&A.

  • State Government

    Application Modernization for State Government: Enhancing Efficiency & Citizen Services

    State governments are increasingly recognizing the need to modernize their legacy applications. With the rapid pace of technological advancements, outdated systems can no longer meet the demands of today’s citizens. Application modernization offers a pathway to more efficient, secure, and user-friendly government services. Why Modernize? Legacy systems often struggle to keep up with the growing […]

  • State Government

    Exploring the Future of AI at the Georgia Emerging Technology Summit: Data & AI 2024

    The Georgia Emerging Technology Summit: Data & AI 2024 was a landmark event for public sector leaders, showcasing the transformative potential of AI and data technologies. This summit brought together key figures and experts to discuss, learn, and network to enhance public service delivery through innovative technology. Keynote Highlights The summit featured insightful keynote sessions […]

  • Data & Analytics

    Legacy Data Modernization: A Comprehensive Guide to Upgrading Your Data Platform

    Though they may have been more than functional in the past, legacy data platforms can become a burden to your organization and prevent it from realizing its full potential. That’s why legacy data modernization can effectively transform your organization’s obsolete data systems into modern platforms that are scalable, efficient, and better equipped to handle today’s […]

  • Strategy & Planning

    7 Trends to Watch in 2025: Leveraging Technology to Achieve Business Objectives

    As we head into 2025, it’s clear that technology isn’t just a part of business strategy — it’s a powerful enabler of success. At UDig, we see it every day: organizations leveraging new tools, technologies and approaches to achieve their goals faster, smarter, and with greater impact. Whether it’s using AI to streamline multiple processes, […]

  • Digital Products

    Energy 2025 – Expansion of Fossil Fuels or Carbon Reduction?

    Now that the election is behind us, we have an opportunity to anticipate the possible effects on the energy industry under this new administration. What strategies will be impacted? What will remain the same? What opportunities can we take advantage of in 2025? This blog is meant to dig into these questions and provide some […]

  • State Government

    Exploring the Future of State Technology: Takeaways from the NASTD 2024 Annual Conference

    Last week, I attended the National Association of State Technology Directors (NASTD) 2024 Annual Conference in Minneapolis, MN. This premier event brings together state technology directors, industry experts, and vendors to discuss the latest trends, challenges, and innovations in state technology. The conference fosters the exchange of ideas and best practices essential for addressing shared […]

  • Data & Analytics

    Masking Data 101: Safeguarding PII in Your Organization

    In today’s digital age, data security and privacy are paramount. As organizations increasingly collect, store, and process personal data, protecting Personally Identifiable Information (PII) has never been more critical. One essential practice that organizations can implement at the database level to secure this sensitive information is to obfuscate it through the usage of data masking […]

  • Insurance

    Insurtech Insights USA 2024: Recap & Key Takeaways

    This was my first time attending Insurtech Insights USA The Insurance Conference 2024. With over 5,000 attendees and 120 sessions with speakers from insurtechs, brokers, carriers, and MGA’s, the event highlighted the industry’s readiness to embrace new tech and solve challenging problems. My colleague, Reid Colson, and I were there to meet with clients and industry […]