Creating a Visual Topic Map for Battery Researchers Using a Large Global Open Dataset

Abstract

Battery related research has gained increasing attention in recent years. A large number of research papers about battery are being published as open access. In order to explore the research content in these papers efficiently, it is necessary to develop information management tools to capture the knowledge embedded in them.We are working on a project to create a visual topic map for battery researchers. This map connects battery researchers according to their research topics. This allows potential connection between researches with similar research topics, which increase collaboration opportunities to advance battery research. Additionally, visualization of research content supports intuitive capturing of research trend by maximizing information absorbance.We extract all published works related to battery research from OpenAlex1, which is a fully open catalog of the global research that offers significant advantages in terms of data coverage and affordability and openness. Anther important feature of OpenAlex is that, every article is tagged with multiple concepts using automatic classifier. In addition to concepts extracted from OpenAlex dataset, we used a pre-trained large language model, named KeyBERT2, to further extract representative terms from the work abstract. KeyBERT is a minimal and easy-to-use keyword extraction technique based on the transformers model.For each researcher, we have extracted all concepts and terms from all the articles that authored by that author. Using those terms and concepts, we formulate a vector for each researcher that represents the total research output. To find researchers with similar research interests, we measure similarity between the vectors in the authors space. Top matching authors are connected visually to create a map expands over representative topics in battery related research. Figure 1 shows a snapshot of this map. Researchers names were hidden for privacy purposes. 1Priem, J., Piwowar, H., & Orr, R. (2022). OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts. ArXiv. https://arxiv.org/abs/2205.01833 2Grootendorst, M. (2020). KeyBERT: Minimal keyword extraction with BERT. https://doi.org/10.5281/zenodo.4461265 Figure 1