Making Sense of Numerical Data - Semantic Labelling of Web Tables

Abstract

With the increasing amount of structured data on the web the need to understand and support search over this emerging data space is growing. Adding semantics to structured data can help address existing challenges in data discovery, as it facilitates understanding the values in their context. While there are approaches on how to lift structured data to semantic web formats to enrich it and facilitate discovery, most work to date focuses on textual fields rather than numerical data. In this paper, we propose a two level (row and column based) approach to add semantic meaning to numerical values in tables, called NUMER. We evaluate our approach using a benchmark (NumDB) generated for the purpose of this work. We show the influence of the different levels of analysis on the success of assigning semantic labels to numerical values in tables. Our approach outperforms the state of the art and is less affected by data structure and quality issues such as a small number of entities or deviations in the data.

Dataset search: a survey

“…For instance, [102] deal with the problem that the underlying dataset cannot be exposed, but good summaries may help the user undertake the task of data access. Meanwhile, [93] use annotations to help support searching over data types and entities within a dataset, while [73] provide better labeling for numerical data in tables.…”

“…Then, the Linked Data cycle can then be applied to these descriptions, ultimately enabling the querying of datasets. The main challenge is the generation and maintenance of these descriptions, with some works tackling the problem of extracting specific properties from specific formats, like [104] for extracting spatio-temporal properties, and [74] for identifying the numerical properties in CSV tables.…”

SemTab 2019: Resources to Benchmark Tabular Data to Knowledge Graph Matching Systems

“…Another option is to use a profiling tool such as Loupe [22]. For table generation with numeric columns, refer to [18].…”

An Assessment of Adoption and Quality of Linked Data in European Open Government Data

“…Vocabulary re-use is limited Our results suggest that publishers are having issues finding, using and/or aligning to vocabularies: from the different ways of assigning values to dct:format, through the default parameters of Socrata's csvtordf conversion, to the low usage of the vocabularies defined by the EC. Follow-up actions Considering that lifting from tabular formats appears to be the best way to move forward, EDP will study the feasibility of applying recent research methods in this area [8,1] to find alignments to tabular headers. This could be done by intermediaries such as the EDP, by portals or the publishers themselves.…”

“…Linked Data and the 5-star scheme are at the core of the open data strategy of the European Commission (EC), described in the Public Sector Information (PSI) Directive. This includes investment in the development and promotion of: metadata specifications such as DCAT-AP 5 to describe datasets; catalogs of resources with persistent URIs; 6 , a data portal to host EC data; 7 as well as the European Data Portal (EDP), 8 which provides a single point of access, search and exploration of open government data by various European public institutions. In November 2015, the EDP has started to harvest metadata from all national portals of the 28 EU countries and associated countries, the EC data portal, and a set of other sources such as geospatial portals.…”