Showing 1–5 of 5 results for author: Valentijn, E A

Structure-Tags Improve Text Classification for Scholarly Document Quality Prediction

Authors: Gideon Maillette de Buy Wenniger, Thomas van Dongen, Eleri Aedmaa, Herbert Teun Kruitbosch, Edwin A. Valentijn, Lambert Schomaker

Abstract: Training recurrent neural networks on long texts, in particular scholarly documents, causes problems for learning. While hierarchical attention networks (HANs) are effective in solving these problems, they still lose important information about the structure of the text. To tackle these problems, we propose the use of HANs combined with structure-tags which mark the role of sentences in the docume… ▽ More Training recurrent neural networks on long texts, in particular scholarly documents, causes problems for learning. While hierarchical attention networks (HANs) are effective in solving these problems, they still lose important information about the structure of the text. To tackle these problems, we propose the use of HANs combined with structure-tags which mark the role of sentences in the document. Adding tags to sentences, marking them as corresponding to title, abstract or main body text, yields improvements over the state-of-the-art for scholarly document quality prediction. The proposed system is applied to the task of accept/reject prediction on the PeerRead dataset and compared against a recent BiLSTM-based model and joint textual+visual model as well as against plain HANs. Compared to plain HANs, accuracy increases on all three domains. On the computation and language domain our new model works best overall, and increases accuracy 4.7% over the best literature result. We also obtain improvements when introducing the tags for prediction of the number of citations for 88k scientific publications that we compiled from the Allen AI S2ORC dataset. For our HAN-system with structure-tags we reach 28.5% explained variance, an improvement of 1.8% over our reimplementation of the BiLSTM-based model as well as 1.0% improvement over plain HANs. △ Less

Submitted 17 December, 2020; v1 submitted 30 April, 2020; originally announced May 2020.

Comments: This new version of the paper brings the paper up-to-date with the improved paper, published at the First Workshop on Scholarly Document Processing, at EMNLP 2020. .Additionally, minor corrections were made including addition of color to Figures 1,2. The changes in comparison to the first arXiv version are substantial, including various additional results, and substantial improvements to the text

ACM Class: I.2.7

Journal ref: Proceedings of the First Workshop on Scholarly Document Processing. Association for Computational Linguistics. (2020) 158-167. EMNLP|SDP 2020 https://www.aclweb.org/anthology/2020.sdp-1.18

Query Driven Visualization

Authors: Hugo Buddelmeijer, Edwin A. Valentijn

Abstract: The request driven way of deriving data in Astro-WISE is extended to a query driven way of visualization. This allows scientists to focus on the science they want to perform, because all administration of their data is automated. This can be done over an abstraction layer that enhances control and flexibility for the scientist. The request driven way of deriving data in Astro-WISE is extended to a query driven way of visualization. This allows scientists to focus on the science they want to perform, because all administration of their data is automated. This can be done over an abstraction layer that enhances control and flexibility for the scientist. △ Less

Submitted 29 November, 2011; originally announced November 2011.

Comments: 4 pages, Procedings ADASS XXI, ASP Conference Series

Journal ref: Astronomical Data Analysis Software and Systems XXI, volume 461, year 2012, page 521

Astro-WISE Information System

Authors: Willem-Jan Vriend, Edwin A. Valentijn, Andrey Belikov, Gijs A. Verdoes Kleijn

Abstract: Astro-WISE is a scientific information system for the data processing of optical images. In this paper we review main features of Astro-WISE and describe the current status of the system. Astro-WISE is a scientific information system for the data processing of optical images. In this paper we review main features of Astro-WISE and describe the current status of the system. △ Less

Submitted 28 November, 2011; originally announced November 2011.

Comments: 4 pages, Proc. of ADASS XXI, ASP Conference Series

Automatic Optimized Discovery, Creation and Processing of Astronomical Catalogs

Authors: Hugo Buddelmeijer, Danny Boxhoorn, Edwin A. Valentijn

Abstract: We present the design of a novel way of handling astronomical catalogs in Astro-WISE in order to achieve the scalability required for the data produced by large scale surveys. A high level of automation and abstraction is achieved in order to facilitate interoperation with visualization software for interactive exploration. At the same time flexibility in processing is enhanced and data is shared… ▽ More We present the design of a novel way of handling astronomical catalogs in Astro-WISE in order to achieve the scalability required for the data produced by large scale surveys. A high level of automation and abstraction is achieved in order to facilitate interoperation with visualization software for interactive exploration. At the same time flexibility in processing is enhanced and data is shared implicitly between scientists. This is accomplished by using a data model that primarily stores how catalogs are derived; the contents of the catalogs are only created when necessary and stored only when beneficial for performance. Discovery of existing catalogs and creation of new catalogs is done through the same process by directly requesting the final set of sources (astronomical objects) and attributes (physical properties) that is required, for example from within visualization software. New catalogs are automatically created to provide attributes of sources for which no suitable existing catalogs can be found. These catalogs are defined to contain the new attributes on the largest set of sources the calculation of the attributes is applicable to, facilitating reuse for future data requests. Subsequently, only those parts of the catalogs that are required for the requested end product are actually processed, ensuring scalability. The presented mechanisms primarily determine which catalogs are created and what data has to be processed and stored: the actual processing and storage itself is left to existing functionality of the underlying information system. △ Less

Submitted 16 November, 2011; originally announced November 2011.

Comments: Accepted for publication in topical issue of Experimental Astronomy on Astro-WISE information system

Query Driven Visualization of Astronomical Catalogs

Authors: Hugo Buddelmeijer, Edwin A. Valentijn

Abstract: Interactive visualization of astronomical catalogs requires novel techniques due to the huge volumes and complex structure of the data produced by existing and upcoming astronomical surveys. The creation as well as the disclosure of the catalogs can be handled by data pulling mechanisms. These prevent unnecessary processing and facilitate data sharing by having users request the desired end produc… ▽ More Interactive visualization of astronomical catalogs requires novel techniques due to the huge volumes and complex structure of the data produced by existing and upcoming astronomical surveys. The creation as well as the disclosure of the catalogs can be handled by data pulling mechanisms. These prevent unnecessary processing and facilitate data sharing by having users request the desired end products. In this work we present query driven visualization as a logical continuation of data pulling. Scientists can request catalogs in a declarative way and set process parameters directly from within the visualization. This results in profound interoperation between software with a high level of abstraction. New messages for the Simple Application Messaging Protocol are proposed to achieve this abstraction. Support for these messages are implemented in the Astro-WISE information system and in a set of demonstrational applications. △ Less

Submitted 11 October, 2011; originally announced October 2011.

Comments: Accepted for publication in topical issue of Experimental Astronomy on Astro-WISE information system