Creating knowledge graphs for geographic data on the web

Abstract

Geographic data plays an essential role in various Web, Semantic Web and machine learning applications. OpenStreetMap and knowledge graphs are critical complementary sources of geographic data on the Web. However, data veracity, the lack of integration of geographic and semantic characteristics, and incomplete representations substantially limit the data utility. Verification, enrichment and semantic representation are essential for making geographic data accessible for the Semantic Web and machine learning. This article describes recent approaches we developed to tackle these challenges.

Semantic Web Technologies and Machine Learning: A Conjoined Silver Bullet for Big Data

“…The essence of data acquisition particularly in data driven economy is to generate insight for efficient decision making by human and machine for increased productivity; a process which is either marred or boosted by the veracity (quality-accuracy, exactness) of data deployed for analyses 24 . Consequently, it is critical to guarantee the veracity of data 3,4,8 . v. Value: This is one of the features of data that came with the advent of data mining-also referred to as knowledge discovery on database (KDD)-whose kernel and end product according to Ebietomere and Ekuobase 10 , is "knowledge"often operationally encapsulated as data 8 .…”

“…A fact further bolstered with the crusade to make machines intuitive, autonomous, and as well perform repetitive task on behalf of human. These data are often big data whose veracity should essentially be guaranteed for it to be valuable for efficient and effective decision making 3,4 by human and machine.…”

“…These physiognomies of big data on the one hand can be greatly beneficial as the analyses of such massive data can give great insight for excellent decision making [6][7][8] -as obvious in many fields or domains including biological sciences, health, law, agriculture, and manufacturing-particularly, with the emergence of industry 4.0 compelling machine to machine interaction 2,3,7,8 . On the other hand, these features birthed several difficulties, pressing among which are; interoperability, and integration [4][5][6]8,9 which have not only rendered the traditional data processing mechanisms impotent but have as well led to the conception, birth, proliferation, and exploration of several disruptive stratagems and technologies. In this report, some of the important technologies revolutionising the acquisition, organisation, analysis, and management of big data are exposed.…”