|Compilers' Toolbox™||- provided by Danish Food Informatics|
Interchange of Food Composition Data
Interchange of food composition data in a broader sense has taken place - in a maybe simple way - ever since the first food composition tables were published. Authors were and are still borrowing data from each other, using and reusing data from different sources, often with the errors and mistakes as the result.
In the more modern sense, food composition data interchange can be regarded as moving the data via electronic means, i.e. electronic storage facilities, networks, etc.
The most important break-through came in the 1980s and 1990s with the developments of computerised food composition databases and the following possibility for electectronic interchange of data.
Today, two main types of interchange can be distinguished. The
first is exchange between FCDMSs Food Composition Data Management
Systems being used for the compilation and evaluation of data
collections - including the supply of data stored in LIMSs
(Laboratory Information Management Systems) by analytical
laboratories and food manufacturers. The second is the
dissemination, often publication, of evaluated data sets.
Exchange between FCDMSs should provide the potential to interchange any data capable of being stored in such a system. FCDM managers must be able to decide what data is appropriately exchanged in a given circumstance and have the facility to tailor the overall interchange specification to that requirement. Since the overall format must be able to accommodate any FCDMS data, logical data design for the FDBMS and for interchange can be considered to be a single operation.
The first report of systematic interchange of food composition
table data was of work undertaken in the Nordic Countries under the
auspices of NORFOODS and supported by the Nordic Council of
A significant portion of the work involved the mechanics of
interchange appropriate to the hardware used at the time in the
various Nordic countries, in practice no longer an issue. The
project compared the data files available for each country's food
tables and made proposals for a minimum standard for food data
interchange. This used a simple approach in which the data file was
separated from the documentation file to the extent that components
were not identified in the data files, values being recognised by
their field position. This close adherence to the layout of printed
food tables might now be considered over-restrictive for general
Indeed at a more recent CEC-funded workshop under the Eurofoods auspecies, it was suggested that it would be useful to review the NORFOODS work in the light of more recent technological developments.
The INFOODS Food Composition Data Interchange Handbook:
forms the basis of the more recent developments (like in Eurofoods
and EuroFIR projects)
in food composition data interchange. It defines the organising principles and formats to be used when
transferring data, for example between regional centres. The report
introduces and and gives an overview covering some basic
considerations which are relevant to the wider adoption of the
The INFOODS format is based on SGML (Standard Generalised Markup Language) - the predecessor of XML (eXtensible Markup Language) - and a clear, unabiguous description of foods and components. The component description is based on the INFOODS publication:
The report proposes how the INFOODS system relates to FCDMS requirements for structuring and handling data, the benefits of using an SGML-based format and their relevance to FDBMS development. and recommends a coordinated approach in defining the specifications for FCDMSs and an interchange format.
|International Interface Standard|
The development of the International Interface Standard (IIS) for the U.S. Food and Drug Administration (Douglass et al., 1995) is closely related to data interchange in that it provides an environment into which interchange files can be imported
In this US system, items can be searched on specific criteria and the associated data retrieved. Food description based on the Langual system provides one of the main retrieval options. The IIS system serves as a reminder that the purpose of interchange is to move useable data from one place to another. Generally, and increasingly, this is between computer facilities managing collections of data. These facilities are frequently supported by FCDMSs which thus need the means to generate interchange files for export and to import them so that the incoming data are correctly interpreted and stored by the receiving FCDMS.
|COST Action 99 - Eurofoods|
The COST Action 99 - Eurofoods (1995-1999) further developed the basics of food composition data management and data interchange. An overview of Data Interchange formats as import/exports formats was published in 1996:
The relationship between interchange formats and FCDMS was
considered further in the COST Action 99 - Eurofoods project
with developments of food composition data structures
and data interchange proposals:
To promote and encourage interchange of food composition data, the COST Action 99 - Eurofoods working
group on food composition data management and interchange proposes a set of recommendations
for data management and interchange using electronic media:
The recommendations are
firmly founded on previous work done internationally by INFOODS and by national agencies
and institutes as well as international standards. The recommendations include guidelines for the
description of foods, components, compositional values and data sources. A sufficiently generic
conceptual schema for food composition is defined to handle food composition data at various
levels of aggregation and with various levels of additional descriptive information.
The recommendations also include technical issues such as file formats and media for data interchange. Furthermore, software tools are presented to assist with implementation of the recommendations.
The Eurofoods recommendations themselves are the first step in a two step approach including a minimum set of requirements for food composition data interchange.
The Eurofoods requirements outline the main categories of data and their description of further data (metadata). The recommendations for this first step proposed a text based interchange format and media for data transfer. A subset of the data information described in the Eurofoods recommendations first step have been used in the data documentation and data transfer from the national compilers of the ten EPIC countries in the EPIC study (European Prospective Investigation into Cancer and Nutrition) coordinated by WHO IARC, Lyon.
The Eurofoods recommendations mention that the second step in the
two step approach could be based on the concepts of SGML (ISO
1986), more recently developed as
XML by the World
Wide Web Consortium in 2000. This
approach has been taken up in the CEECFOODS (the sub-regional food
composition network for the Central and Eastern European States) and
implemented in the Alimenta and the Data Center and Data Management
System software developed by FloraFood (websites/references no
The software allowed for interchange of food composition data using a so-called TransportPackage in XML. This is the first attempt to use an open standard XML format to perform food composition data interchange between partners (in this case, the CEECFOODS network) and internationally. It represents a pragmatic and feasible system that works. The TransportPackage was presented at the International Food Data Conference in Bratislava in 2001.
|FAO Technical workshop on Standards for food composition data interchange|
The Technical workshop on Standards for food composition data interchange, held in Rome, 19-22 January 2004 "noted that XML would be the most suitable way to interchange data but that technical limitations exist which make it difficult for some compilers, including those in North America and Europe, to convert their data into XML. It was therefore agreed that the next step towards the XML interchange would be a set of relational files with standardized tags and definitions which would then be converted into XML format guidelines. These files should therefore be regarded as collections of elements and their attributes that can be presented in a single file or as a set of relational files". This approach is similar to the two step approach of the Eurofoods recommendations, but introduces a more elaborated, theoretical concept of definitions and relationships than implemented in the Eurofoods scheme.
|EuroFIR Network of Excellence|
The EuroFIR Network of Excellence (2006-2010) was a five-year
Network of Excellence funded by the European Commission's Research
Directorate General under the "Food Quality and Safety Priority" of
the Sixth Framework Programme for Research and Technological
Development. The network involved 49 partners from universities,
research institutes and small-to-medium sized enterprises from 27
In the EuroFIR project, one of the major tasks was to set up – and implement - the EuroFIR Databank System, the EuroFIR eSearch facility, now replaced by EuroFIR FoodEXplorer.
The constructed data retrieval facilities allow users to specify foods and components, return relevant data, and provide quality measures of the retrieved data matrix. The EuroFIR specifications for a European food composition databank system are based on the actual availability of data and metadata and use and implement an updated subset of the Eurofoods recommendations. The Eurofoods recommendations form the background of definitions and concepts used in the EuroFIR specifications. Whenever possible, the EuroFIR specifications take into account international recommendations and cross-references between local, national and international concepts and entities will be given.
To underpin the EuroFIR NoE specifications, a series of
documentation reports concerning data structures and interchange
facilities have been published:
The development of
the system, which allows easy access and interchange of European and
other food composition data, is build on state-of-the-art techniques
using XML (eXtensible Markup Language) and REST/SOAP inplementations
to transport data between local and cetral servers. The resources
are build on the comprehensive value documentation developed by the
EuroFIR projects and facilitates the retrieval and use of information on foods, food
components, calculation parameters, analytical methods, source
references and other food-related topics.
For more information on EuroFIR Thesauri and Value Documentation, see Value Documentation.
In connection with the development of the EuroFIR Food Data Transport Package and Meta Data Transport Package XML templates and the EuroFIR Web Services, a series of XML schemata (schemas) were developed to support the integrity and structure of the XML template:
The Schemata documentations include definitions of the special
data types required in food data interchange. An example is the
values that should not be defined as a numbers (decimal or integer),
which restrict the values being interchanged to real numbers, and
not the often appearing “no value” (an empty string, used in EuroFIR
when no Selected value (mandatory) can be assigned when extreme
variability is characterizing the value and only a minimum and a
maximum can be assigned) or “0-1” (a string often used as a value in
nutrition labelling to indicate that the value is lower that one,
but bigger than zero). The value type chosen by EuroFIR is
"decimal-as-string", which preserves the precision of a number
(significat digits) and at the same time can be used for string
information, like the empty string (no value) or value expressed as
For more details on values in numerical databanks, see Significant Digits.
Apart from the EuroFIR thesauri mentioned above, the EuroFIR Food Data Transport Package template makes use of several of other thesauri:
In addition to language itself, it is often important to distinguish between dialects of a language, e.g. British English and American English. Although the ISO 639-6 standard include ways of indicating languages used in different countries/regions, it was decided to use the current best practice as described in Internet Society's RFC 4646 and RFC 4647 (Tags for the Identification of Languages).
In these documents, the Internet Society describes the structure, content, construction, and semantics of language tags in a faceted approach for use in cases where it is desirable to indicate the language used in an information object in Internet applications. It also describes how to register values for use in language tags and the creation of user-defined extensions for private interchange.
The language tag consists of a primary sub-tag and a series of subsequent sub-tags, each of which narrows or refines the range of languages identified by the overall tag. It enables the user to specify, in addition to the primary language, other characteristics such as script, country, or variant.
It is considered an Internet Best Current Practices
for the Internet Community and gives guidance for the use of ISO
RFC 4646 specifies use of a 2-character code from ISO
639-1 when it exists; when a language does not have a
2-character code assigned the 3-character code is used. Although
it states that the 3-character terminology code is used in these
cases where no 2-character code exists, this situation will not
occur, since the only alternative codes in ISO 639-2 are for
languages that already have a 2-character code.
Some (simple) examples are
The specifications for the EuroFIR Web Services developed during the EuroFIR project were futher developed during the two year project extension named EuroFIR - Nexus (2011-2013). The following reports were published:
|CEN Standard for food data - EN 16104:2012|
As a spin-off of the EuroFIR projects, a project committee under the European Standardisation auspices, the CEN/TC 387 food data project committee, was initiated. The work of the CEN/TC 387 finalised its work in March 2012, agreed upon by a final vote by the CEN member countries in August 2012 and published as a European Standard, EN 16104:2012, Food data - Structure and interchange format, on 3 November 2012 (preview).
The European Standard specifies requirements on the structure and
semantics of food datasets and of interchange of food data for
various applications. Food data refers to information on various
food properties and includes various steps in the generation and
publication of such data, e.g. sampling, analysis, food description,
food property and value description.
The standard regards food data as datasets covering:
The standard includes requirements on:
The standard does not include: