Recent developments in the field of data mining (DM) have received considerable attention from the materials science community due to their ability to accelerate the design of new materials. Experimental datasets of materials are usually published in scientific literature. Mining such literature thus enables the possibility of discovering synergistic effects and meaningful insights by the virtue of evaluating the combined experimental datasets. Therefore, the selection of relevant data is essential for DM. Unfortunately, such data is not easily found using provided search features of digital libraries, such as Scopus, ScienceDirect, arXiv, and Europe PMC. Highly specific searches, such as retrieval of literature that has exclusively TiAl-Creep experimental datasets, are not possible in these digital libraries.

This research work presents a system that facilitates a generic search-based ingestion of literature from digital libraries, followed by the selection of DM relevant literature. Besides phrase, facet, full-text, and conjunctive and disjunctive search capabilities, the selection mechanism also allows dataset-aware literature retrieval through figure caption, figure type and characteristics, and domain knowledge taxonomy-based semantic searches. This allows finding experimental datasets that are represented as visual elements embedded within literature. Furthermore, the system employs a matching-based approach to produce concise summaries of search results. The flexibility of the proposed system also enables it to be seamlessly applied to other domains for literature mining purposes, opening up a multitude of research opportunities. This poster attempts to address the importance of such a system by highlighting its features and applicability in a materials science domain.