ForesightX

Inspiration

Universities invest millions of dollars in research, but how do they know where to focus next? Traditional metrics, such as citation or publication count, show what’s already popular, not what’s emerging.

What it does

Foresight X is a student-built research intelligence platform that allows Boston University to make data-driven decisions about future research investments. It combines academic data, social signals, and institutional performance to forecast where BU’s next big breakthrough should happen. The platform evaluates every research topic using four customizable metrics: Publication Momentum — captures how fast a research field is growing in academia using publication data and compound annual growth rate (CAGR).

BU Gap Score — identifies where BU is underrepresented compared to global research activity, spotlighting potential strategic investment opportunities.

Interdisciplinary Spread — quantifies how cross-cutting a topic is across academic domains, helping BU find collaboration-rich frontiers.

Social Velocity — measures how much a research topic is gaining public and online traction, revealing “zeitgeist” influence from social media and news.

Users can adjust the weights of each metric to fit their strategic priorities — for instance: Increase Gap Score weight to find where BU should catch up. Emphasize Social Velocity to chase hot emerging trends. Prioritize Interdisciplinary Spread to foster collaborative innovation. Foresight X helps Boston University: Stay ahead of the curve in emerging science and technology. Allocate research funding strategically toward high-growth areas. Foster cross-disciplinary innovation across departments. Bridge academic insight with public interest, ensuring research remains relevant to society.

This makes Foresight X not just a static dashboard but a dynamic foresight tool for BU’s leadership, research offices, and even faculty strategizing grant proposals.

How we built it

Data Consolidation Queried OpenAlex for publication and preprint counts by topic from 2021–2025. Cleaned and merged datasets to create a year-by-year trend table for over 250 research topics. Extracted BU-specific publication counts to measure the university’s relative representation in each field.

1. Metric Engineering Publication Momentum Computed Compound Annual Growth Rate (CAGR) for each topic’s publication trend:

$$ {CAGR} = \left(\frac{N_{2025}}{N_{2021}}\right)^{1/4} - 1$$ Identified top-growth topics (e.g., AI in Healthcare, Privacy-Preserving Technologies, FinTech).

Normalized values to allow fair comparison across disciplines.

BU Gap Score Compared BU’s share of global publications in each topic: $$ \text{Gap Score} = 1 - \frac{\text{BU Publications}}{\text{Global Publications}}$$ High Gap Score = major growth opportunity for BU.

Interdisciplinary Spread Retrieved each topic’s associated concept hierarchy from OpenAlex (e.g., belongs to AI, Medicine, Education). Computed the proportion of distinct parent disciplines per topic: $$ S = \frac{n_{\text{linked disciplines}}}{n_{\text{total possible}}}​​ $$ Higher S means the field bridges multiple domains → high potential for collaboration.

Social Velocity Used Crossref’s Event Data API to track DOI mentions on social media, blogs, and news.

Calculated normalized growth in mentions for 2024–2025: $$ V = z\text{-score}(\text{Social Engagement}) $$ Captures how public attention toward a field relative to average attention.

1. Composite Foresight Index We integrated all four metrics into a weighted composite score:

$$ {Foresight Index} = w_1 \cdot M + w_2 \cdot G + w_3 \cdot S + w_4 \cdot V $$ where M,G,S,V represent the normalized scores of each metric, and the wi​ weights are user-customizable via sliders on the platform interface.

Platform Prototype Built with Streamlit, the dashboard allows users to: Visualize research momentum as trend lines and growth curves. Adjust weights dynamically to recalculate forecasts in real time. Highlight “High Impact – High Gap” topics where BU should strategically focus. Export customized foresight reports to Excel or PDF for administrative use.

Challenges we ran into

OpenAlex data overload With millions of publications, extracting and processing data was computationally intensive and time-consuming. To address this, we used the compound annual growth rate to identify the 15 most trending research topics and focused our data extraction on those areas. This approach tremendously improved time and computational efficiency.

Social media API restrictions PRAW (Python Reddit API Wrapper) has strict regulations on data scraping to protect user privacy, and we chose to ethically respect these limitations. Additionally, Tweepy and other Twitter scraping libraries are no longer supported, blocking our initial approach. We turned to Crossref Event Data, which conveniently tracks how many times DOIs, or research papers, have been mentioned in social media posts, enabling us to track engagement through that.

Accomplishments that we're proud of

APP IS LIVE! and interactive while planning we used relational schema (shout out to my DS310 professor!) all of our metrics relied on statistical methods, which is a lot more math heavy than anything we’ve dne before.

What we learned

We gained valuable experience working with OpenAlex, a large-scale research database, while building a professional interactive dashboard with Streamlit. The core challenge was developing an algorithm that synthesizes four metrics into a single, interpretable opportunity score that guides strategic research decisions.

What's next for ForesightX

Our next steps include creating a comparison page showing socially trending topics versus academically trending topics to identify overhyped research areas and reveal genuine breakthrough opportunities. We’ll also implement live data feeds from OpenAlex and Crossref Event Data, providing real-time updates to the opportunity rankings.

Built With

• crossref-doi-resolver
• openalex
• python
• streamlit