Research Article
A Secured Preposition-Enabled Natural Language Parser for Extracting Spatial Context from Unstructured Data
@INPROCEEDINGS{10.1007/978-3-319-98827-6_14, author={Patience Usip and Moses Ekpenyong and James Nwachukwu}, title={A Secured Preposition-Enabled Natural Language Parser for Extracting Spatial Context from Unstructured Data}, proceedings={e-Infrastructure and e-Services for Developing Countries. 9th International Conference, AFRICOMM 2017, Lagos, Nigeria, December 11-12, 2017, Proceedings}, proceedings_a={AFRICOMM}, year={2018}, month={8}, keywords={Knowledge representation Ontology Spatial reasoning Unstructured data}, doi={10.1007/978-3-319-98827-6_14} }- Patience Usip
Moses Ekpenyong
James Nwachukwu
Year: 2018
A Secured Preposition-Enabled Natural Language Parser for Extracting Spatial Context from Unstructured Data
AFRICOMM
Springer
DOI: 10.1007/978-3-319-98827-6_14
Abstract
Acquiring data within the health domain is generally intractable due to privacy or confidentiality concerns. Given the spatial nature of health information, and coupled with the accompanying large and unstructured dataset, research in this area is yet to flourish. Further, obtaining spatial information from unstructured data is very challenging and requires spatial reasoning. Hence, this paper proposes a secure Preposition-enabled Natural Language Parser (PeNLP), sufficient for mining unstructured data to extract suitable spatial reference with geographic locations. The proposed PeNLP is a subcomponent of a larger framework: the Preposition-enabled Spatial ONTology (PeSONT) – an ongoing project. The short term impact of PeNLP is its availability as a reliable information extractor for spatial data analysis of health records. In the long run, PeSONT shall aid quality decision making and drive robust policy enactment that will greatly impact the health sector and the populace.
