MatScIE: An automated tool for the generation of databases of methods and parameters used in the computational materials science literature

An automated materials and processes identification tool for material informatics using deep learning approach

“…Besides these studies, some studies in the field of materials science have also been carried out by [19] , [20] , [21] , [15] , [13] . All the studies that focus on the field of materials science have one thing in common: they focus on specific materials or only on the synthesis methods.…”

“…All the studies that focus on the field of materials science have one thing in common: they focus on specific materials or only on the synthesis methods. In the work of Guha et al [20] they extracted five different types of information, such as materials, processes, parameters, formulas, and structures, from the scientific literature in materials science. They are only interested in specific materials and related information mentioned in the title or system user input.…”

“…Moreover, the proposed tool can extract associated values for the entities and represent the entire knowledge in a graph-based knowledge map. The proposed tool also uses the Bi-LSTM layer within the deep neural network model like the studies mentioned in [19] , [20] , [13] . Among all the works mentioned, the proposed one received the highest F1 score of 96% whereas [19] achieved 87%, and [20] achieved 73.13% F1 score.…”

“… • Uses multiple machine learning models but achieves lower F1 scores than existing systems. [20] • This system employed an LSTM-based neural network for the NER task. • Uses a comparatively large amount of training data.…”

Sentence Boundary Extraction from Scientific Literature of Electric Double Layer Capacitor Domain: Tools and Techniques

“…Complete sentence recognition plays an important role in an automatic text processing system because a complete sentence contains both syntactic and semantic information. Different studies have used different tools and techniques for PDF extraction and sentence boundary detection in the creation of different chemical or material information systems, without providing a rationale for their choice of tools and the performance of the tools for PDF extraction and sentence boundary detection [36][37][38]. Therefore, a comparison between the existing PDF and text extraction tools is required to find a suitable one.…”