Fonduer: Knowledge Base Construction from Richly Formatted Data
By Sen Wu, Luke Hsiao, Xiaoxia Cheng, Braden Hancock, Theodoros Rekatsinas, P. Levis, Christopher R'e. 2017
99 citations recorded by the source corpus.
We focus on knowledge base construction (KBC) from richly formatted data. In contrast to KBC from text or tabular data, KBC from richly formatted data aims to extract relations conveyed jointly via textual, structural, tabular, and visual expressions. We introduce Fonduer, a machine-learning-based KBC system for richly formatted data. Fonduer presents a new data model that accounts for three challenging characteristics of richly formatted data: (1) prevalent document-level relations, (2) multimodality, and (3) data variety. Fonduer uses a new deep-learning model to automatically capture the representation (i.e., features) needed to learn how to extract relations from richly formatted data. Finally, Fonduer provides a new programming model that enables users to convert domain expertise, based on multiple modalities of information, to meaningful signals of supervision for training a KBC system. Fonduer-based KBC systems are in production for a range of use cases, including at a major online retailer. We compare Fonduer against state-of-the-art KBC approaches in four different domains. We show that Fonduer achieves an average improvement of 41 F1 points on the quality of the output knowledge base---and in some cases produces up to 1.87x the number of correct entries---compared to expert-curated public knowledge bases. We also conduct a user study to assess the usability of Fonduer's new programming model. We show that after using Fonduer for only 30 minutes, non-domain experts are able to design KBC systems that achieve on average 23 F1 points higher quality than traditional machine-learning-based KBC approaches.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Christopher R'e
- Paroma Varma
- Zongben Xu
- Jun Shu
- Theodoros Rekatsinas
- Deyu Meng
- Ganesh Krishnan
- Han Li
Citations in the corpus
- Small Sample Learning in Big Data Era (2018, background): “After the raising of this idea, Wu et al. (2018) proposed Fonduer, together with data programming, to provide weak supervision of domain expertise to guide extract information from richly formatted data.”
- Snuba: Automating Weak Supervision to Label Training Data (2018, methodology): “These primitives are user-defined and part of open source libraries [36, 51] and data models in existing weak supervision frameworks [39, 59].”
- Deep Learning for Entity Matching: A Design Space Exploration (2018, methodology): “What makes this challenge unique to EM is that existing weak supervision approaches have focused primarily on textual data for tasks such as information extraction [61, 85], or visual data for tasks such as image classification [54].”