Home Knowledge Mining Big Data

Mining Big Data

 

As discussed in our previous article here the Department of Business, Enterprise and Innovation has put forward the Copyright and Other Intellectual Property Law Provisions Bill 2018 (the “Bill”) which seeks to make a number of amendments to the Irish Copyright and Related Rights Act 2000 (the “Copyright Act”).  

One of the most notable amendments being proposed is the exception to copyright infringement for text and data mining. This exception was one of the recommendations made in the ‘Modernising Copyright’ report issued by the Copyright Review Committee (“CRC”). However, the wording of the provision is quite different to that proposed by the CRC. 

In their report, the CRC noted that “very significant social benefits stand to be gained from text-mining and data-mining” given the potential for new discoveries from existing data. 

What is data mining? 

Data mining is essentially the process of analysing large data sets in order to try and identify patterns, trends, anomalies and establish relationships. 

‘Text and data mining’ was defined in the draft EU Directive on Copyright in the Digital Single Market (the “draft EU Directive”) as meaning “any automated analytical technique aiming to analyse text and data in a digital form in order to generate information such as patterns, trends and correlations”. 

We are now in the era of Big Data with an unprecedented amount of new data being created every day. Researchers cannot keep up with the level of data being created and therefore need to rely on technology to assist in analysing the data. Further as data mining identifies patterns, it serves as a foundation for Artificial Intelligence and machine learning. However, the existing exceptions under the Copyright Act do not provide protection for some of the activities being conducted by the relevant technologies. For instance, when large data sets are being analysed, copies of the data (a large proportion of which may be protected by copyright) are usually created/reproduced by the relevant software to allow the process to work. Often such copies are not temporary and may have an independent economic significance and therefore the exceptions to copyright infringement under the Infosoc Directive cannot be utilised. 

What does the Bill propose?

The new Section 53 A provides:

  • Subject to subsection (3), the making of a copy of a work by a person who has lawful access to the work does not infringe the copyright in the work where the copy is –
    • made in order that the person may carry out a computational analysis of anything in the work for the sole purpose of research for a non-commercial purpose, and
    • is accompanied by a sufficient acknowledgement.

Therefore, the exception is only available:

  • to a user who had lawful access to the work – It does not confer a right of access to a work where none existed;
  • the user must be using the work solely for the purpose of research for a non-commercial purpose;
  • there must be sufficient acknowledgement.

Section 53 A (2) goes on to provide that if the copy was made when the work was available without restriction, then it does not matter if access to the work is later restricted.

Section 53 A (3) puts in place safeguards regarding further use of copies made under the data mining exception.  The copyright in the work is infringed where such copy is 

  • transferred to any other person, except where the transfer is authorised by the copyright owner, or
  • is used for any purpose other than the purpose referred to in subsection 1 (a).

Section 53 A (5) provides that where the publication of the results of the computational analysis include the reproduction of extracts from a work, such inclusion shall constitute inclusion in an incidental manner if the extracts are not more than reasonably necessary to explain or to assist in explaining the results of the analysis.

It is important to note that Section 2 (10) of the Copyright Act already renders redundant any terms or condition in an agreement which try to override or restrict any of the legislative exceptions. 

Bill is wider than exception currently proposed at EU level 

The data mining exception proposed in the Bill is actually wider than that proposed by the EU Commission in September 2016 in the draft EU Directive (referred to above) wherein the data mining exception was only available to research organisations.  

Research organisation was defined in that draft EU Directive as meaning “a university, a research institute or any other organisation the primary goal of which is to conduct scientific research or to conduct scientific research and provide educational services: (a) on a non-for-profit basis or by reinvesting all of the profits in its scientific research; or (b) pursuant to a public interest mission recognised by a Member State”.

Whilst the recitals to the draft EU Directive did not rule out public-private partnerships being able to avail of the exemption, it did provide that the exemption would not be available to such organisations where commercial undertakings have decisive influence or control over them.

The proposed data mining exception is a very welcome development which should assist in facilitating research inevitably leading to important and lifesaving discoveries in vast array of areas including medicine, sciences and technology. 

Contributed by Colette Brady.

 

Twitter

 

Followed us @WilliamFryLaw

 

Back to Legal News