online mining boom?

Following on from Nigel’s thoughts on personalisation and data mining, I have some of my own.

I have traditionally regarded the core purpose of an LMS to complete daily operational tasks such as accessioning a magazine, recording borrower activity, keeping track of stock, and a bit of collection profiling i.e. see what is being borrowed and what is not.

But with increasing data mining capabilities in library system there is a potential wealth of info to be mined from patron interactions with the library and each other. Sorcer has given us a glimpse of this potential, in “Books for you” and “Others have read”.

Data mining is not necessarily new to libraries, for example most if not all libraries are required to mine information for their annual reports.  All anonymous of course, and for internal reasons such as justifying budgets and meeting KPIs.

However, if we take the case of commercial ventures such as Facebook, Google, Foursquare et. al. , our online activities are much more closely scrutinised, being used for targeted advertising, collecting market research data, suggesting new friends and contacts etc.

This raises an interesting question, can libraries do the same thing? For example, interrogating the LMS to find out about borrowing patterns could also have many uses, commercial and social. There are of course potential ethical and security considerations, but nonetheless gives us food for thought and so I’ve listed some specific questions below.

Data mining for commercial gain
Can a library sell its data to outside organisations for marketing/retail purposes? Have any libraries done this? Would libraries ever do this to get funding to keep the library operating or add new services?

Ethical and security considerations
Are libraries breaking confidentially agreements by providing information about certain groups of users to external parties?  Can we be confident that data is adequately anonymised so that only demographic trends are recorded rather than individual movements?

Who would manage the process
Internal staff, managed services from vendor, or third parties who may not even be library related?

I should add that I’m not necessarily for or against any of these ideas, simply curious.

About Emerald

Emerald Leung, Sales & Marketing Manager for Civica LLD based in Singapore.
This entry was posted in Libraries and tagged , , , . Bookmark the permalink.

4 Responses to online mining boom?

  1. Nina Antosz says:

    Would any organization be interested in buying library data? Bookshops are dying and libraries have to change to survive.
    Currently, public still see the library as the safe and trustworthy place. Wouldn’t we lost this image when we will start to sell the data, we probably would lost a considerable number of members as they will not trust us.
    Interesting concept though.

  2. Hugh says:

    For me, I think selling data is something no public library can really consider. Our number one asset, far more valuable than anything else, is trust. Our communities and patrons trust us to provide objective and useful advice, but also to keep their personal information confidential and secure. It’s not just a matter of keeping it anonymised – we have to remain beyond suspicion.

    Having said that, I think libraries (and vendors like Civica) can certainly do more in terms of using the huge trove of data that is available. One example I’ve been thinking about (and meaning to ask you, Emerald!) is the possibility of collecting and using data on what patrons are searching for, and clicking on, when they search our catalogue in the same way that Google collects such data. This could be used in a variety of ways, including relevance ranking for search results, but the thing I’ve been considering is using it as a guide for collection development. We may find, for example, that our patrons have been doing a lot of searches for a subject where our collection is old or weak, or a title we don’t hold.

  3. Nigel says:

    For me the big question is the transparency of the commercialisation process. As a consumer I am uneasy in the way in which search engines in general and Google in particular manipulate the advertisements I see and the search results I get as a consequence of my browsing habits outside the chosen search engine.

    Fair enough that Google learn that when I search for harrow I am more likely to be interested in the farm implement than the school or experiences and deliver advertisements on that basis and rearrange the search results to put agricultural suppliers at the top of the list. However the fact that I have searched for harrow in my library’s OPAC is between me and my library!

  4. Nina Antosz says:

    I have to agree on that one with Nigel. I am constantly asked for OPAC stats by using similar techniques as Google analytics, but in my opinion it’s a big invasion of privacy. Maybe I am a bit paranoid coming from ex-communist country, don’t want my search to be controlled and scrutinized.
    Hugh you said that “It’s not just a matter of keeping it anonymised – we have to remain beyond suspicion”, I am not sure if library users will be happy if they will know that library can analyze their searchers.
    As for collection development we can use borrowing stats, in-house usage, databases usage and web analytics.

Comments are closed.