Browsing by Author "Jayanetti, Himarsha R."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Automatic Metadata Extraction Incorporating Visual Features from Scanned Electronic Theses and Dissertations
    Choudhury, Muntabir; Jayanetti, Himarsha R.; Wu, Jian; Ingram, William A.; Fox, Edward (IEEE, 2021-09-27)
    Electronic Theses and Dissertations (ETDs) contain domain knowledge that can be used for many digital library tasks, such as analyzing citation networks and predicting research trends. Automatic metadata extraction is important to build scalable digital library search engines. Most existing methods are designed for born-digital documents such as GROBID, CERMINE, and ParsCit, so they often fail to extract metadata from scanned documents such as for ETDs. Traditional sequence tagging methods mainly rely on text-based features. In this paper, we propose a conditional random field (CRF) model that combines text-based and visual features. To verify the robustness of our model, we extended an existing corpus and created a new ground truth corpus consisting of 500 ETD cover pages with human validated metadata. Our experiments show that CRF with visual features outperformed both a heuristic baseline and a CRF model with only text-based features. The proposed model achieved 81.3%-96% F1 measure on seven metadata fields. The data and source code are publicly available on Google Drive1 and a GitHub repository2.
  • Thumbnail Image
    MetaEnhance: Metadata Quality Improvement for Electronic Theses and Dissertations of University Libraries
    Choudhury, Muntabir Hasan; Salsabil, Lamia; Jayanetti, Himarsha R.; Wu, Jian; Ingram, William A.; Fox, Edward A. (ACM, 2023)
    Metadata quality is crucial for discovering digital objects through digital library (DL) interfaces. However, due to various reasons, the metadata of digital objects often exhibits incomplete, inconsistent, and incorrect values. We investigate methods to automatically detect, correct, and canonicalize scholarly metadata, using seven key fields of electronic theses and dissertations (ETDs) as a case study. We propose MetaEnhance, a framework that utilizes state-of-the-art artificial intelligence (AI) methods to improve the quality of these fields. To evaluate MetaEnhance, we compiled a metadata quality evaluation benchmark containing 500 ETDs, by combining subsets sampled using multiple criteria. We evaluated MetaEnhance against this benchmark and found that the proposed methods achieved nearly perfect F1-scores in detecting errors and F1-scores ranging from 0.85 to 1.00 for correcting five of seven key metadata fields. The codes and data are publicly available on GitHub11https://github.com/lamps-lab/ETDMiner/tree/master/metadata-correction.