Information Extraction Based on Multi-turn Question Answering for Analyzing Korean Research Trends

Analyzing Research and Development(R&D)trends is important because it can influence future decisions regarding R&D direction.In typical trend analysis,topic or technology taxonomies are employed to compute the popularities of the topics or codes over time.Although it is simple and effective,the taxonomies are difficult to manage because new technologies are introduced rapidly.Therefore,recent studies exploit deep learning to extract pre-defined targets such as problems and solutions.Based on the recent advances in question answering(QA)using deep learning,we adopt a multi-turn QA model to extract problems and solutions from Korean R&D reports.With the previous research,we use the reports directly and analyze the difficulties in handling them using QA style on Information Extraction(IE)for sentence-level benchmark dataset.After investigating the characteristics of Korean R&D,we propose a model to deal with multiple and repeated appearances of targets in the reports.Accordingly,we propose a model that includes an algorithm with two novel modules and a prompt.A newly proposed methodology focuses on reformulating a question without a static template or pre-defined knowledge.We show the effectiveness of the proposed model using a Korean R&D report dataset that we constructed and presented an in-depth analysis of the benefits of the multi-turn QA model.

LAME:Layout-Aware Metadata Extraction Approach for Research Articles

The volume of academic literature,such as academic conference papers and journals,has increased rapidly worldwide,and research on metadata extraction is ongoing.However,high-performing metadata extraction is still challenging due to diverse layout formats according to journal publishers.To accommodate the diversity of the layouts of academic journals,we propose a novel LAyout-aware Metadata Extraction(LAME)framework equipped with the three characteristics(e.g.,design of automatic layout analysis,construction of a large meta-data training set,and implementation of metadata extractor).In the framework,we designed an automatic layout analysis using PDF Miner.Based on the layout analysis,a large volume of metadata-separated training data,including the title,abstract,author name,author affiliated organization,and keywords,were automatically extracted.Moreover,we constructed a pre-trainedmodel,Layout-Meta BERT,to extract the metadata from academic journals with varying layout formats.The experimental results with our metadata extractor exhibited robust performance(Macro-F1,93.27%)in metadata extraction for unseen journals with different layout formats.

Importance of surface oxide for the tribology of a Zr-based metallic glass

Thermally grown surface oxide layers dominate the single-asperity tribological behavior of a Zr60Cu30Al10 glass.Increase in oxidation time leads to an increased contribution of shearing and a corresponding decreased contribution of ploughing to friction.This change in the dominating friction and wear mechanism results in an overall minor decrease of the friction coefficient of oxidized surfaces compared to the metallic glass sample with native surface oxide.Our results demonstrate the importance of creating a stable oxide layer for practical applications of metallic glasses in micro-devices involving sliding contact.

Deformation-and rupture-controlled friction between PDMS and a nanometer-scale SiO_(x) single-asperity

This work investigates the friction between polydimethylsiloxane(PDMS)and silicon oxide(SiO_(x))in single asperity sliding contact by atomic force microscopy(AFM).Two friction dependences on the normal force are identified:a tensile regime and a compressive regime of normal forces.In the compressive regime,friction is governed by the shear deformation and rupture of junctions between PDMS and SiO_(x).In this case,the shear strengthτ≈10 MPa is comparable with the cohesive strength of PDMS under compressive loading.In contrast,friction in the tensile regime is also affected by the elongation of the junctions.The single SiO_(x)-asperity follows a stick–slip motion on PDMS in both normal force regimes.Statistical analysis of stick–slip as a function of the normal force allows determining the necessary amount of energy to break a SiO_(x)/PDMS junction.Friction between a SiO_(x)-asperity and a PDMS surface can be rationalized based on an energy criterion for the deformation and slippage of nanometer-scale junctions.