VizCV: AI-assisted visualization of researchers' publications tracks

TL;DR

VizCV tackles the challenge of understanding a researcher’s career trajectory by integrating topic modeling, scientometrics, and egocentric collaboration visualization into an interactive web platform. It uses an end-to-end pipeline that converts Scopus profiles into topic structures and metrics via $Specter$ embeddings, BERTopic topic modeling, $UMAP$ dimensionality reduction, and $HDBSCAN$ clustering, complemented by an LLM-driven textual report generator. The system provides multi-tab views for topic evolution, publication impact, and co-authorship dynamics, plus comparative and scenario-driven analyses to support recruitment, evaluation, and collaboration decisions. Evaluation with university officials indicates practical utility, though limitations include dependence on abstracts/keywords and token costs for LLM inferences. Future work includes expanding data sources, enabling direct profile uploads, enhancing metrics, and releasing the codebase for broader adoption.

Abstract

Analyzing how the publication records of scientists and research groups have evolved over the years is crucial for assessing their expertise since it can support the management of academic environments by assisting with career planning and evaluation. We introduce VizCV, a novel web-based end-to-end visual analytics framework that enables the interactive exploration of researchers' scientific trajectories. It incorporates AI-assisted analysis and supports automated reporting of career evolution. Our system aims to model career progression through three key dimensions: a) research topic evolution to detect and visualize shifts in scholarly focus over time, b) publication record and the corresponding impact, c) collaboration dynamics depicting the growth and transformation of a researcher's co-authorship network. AI-driven insights provide automated explanations of career transitions, detecting significant shifts in research direction, impact surges, or collaboration expansions. The system also supports comparative analysis between researchers, allowing users to compare topic trajectories and impact growth. Our interactive, multi-tab and multiview system allows for the exploratory analysis of career milestones under different perspectives, such as the most impactful articles, emerging research themes, or obtaining a detailed analysis of the contribution of the researcher in a subfield. The key contributions include AI/ML techniques for: a) topic analysis, b) dimensionality reduction for visualizing patterns and trends, c) the interactive creation of textual descriptions of facets of data through configurable prompt generation and large language models, that include key indicators, to help understanding the career development of individuals or groups.

Figures (9)

• Figure 1: Pipeline of our data processing stack. The Scopus abstracts and keywords are processed through Sentence Transformer, to obtain embeddings for the documents. These embeddings are reduced to 2D for the scatterplot, and to 10D to create clusters using HDBSCAN. Topics are then extracted using BERTopic.
• Figure 2: Evolution of research topics over the last 15 years of a researcher. It is clearly visible how the interest has shifted from Rendering (orange) to Deep Learning (red), and the number of papers published per year has increased (see red bar in the bar chart) recently.
• Figure 3: Comparing two researchers in the main view. The donut charts facilitate understanding of the distribution of publications between the first (lighter) and second (darker) researchers. These tones are used for all the clusters. Here only the last nine years are selected instead of the whole available time range.
• Figure 4: Points visualization with a hover operation that shows details on the hovered point.
• Figure 5: The prompt configuration tool lets the user indicate the metrics that will be present in the report, and automatically calculates the number of tokens (with a threshold of 5K) to ensure the limit is not surpassed.