NBD Biblion MIPS

Machine Learning and Artificial Intelligence Integration

NBD Biblion – MIPS

NBD Biblion offers products, services and expertise in the field of media and content. They are the largest supplier of information and purchase advice for Public Libraries and (school) media centres in the Netherlands. They also supply many of the books available in libraries. Hundreds of new books and other media arrive at NBD Biblion every week. Each title is classified and reviewed by NBD editors and reviewers. As a result, they can provide up-to-date information about every newly published title.

We created the Media Information and Production System (MIPS) used by NBD Biblion. A custom-built production and workflow management system for media titles at the heart of their production processes. NBD Biblion uses MIPS to describe and catalogue hundreds of new titles every month. In particular, the software regulates and monitors production planning, monitoring the progress of each title through the system. In addition, we share the information stored in MIPS with many other NBD applications using APIs.

Since 2018, NBD Biblion has worked with a company specialising in Machine Learning, developing data-driven collection algorithms and models that help NBD Biblion deliver individually tailored weekly ordering advice to their clients. They are also working on AI developments that automatically creates metadata for books and other media titles.

Machine Learning
Machine Learning

Automated metadata using AI

Metadata is a term that may not be familiar to everyone, but metadata refers to characteristics that can describe any object – including books. Metadata can encompass basic details such as the title and author. However, it can also include a wealth of other information, such as the publisher, date of publication, ISBN/EAN number, number of pages, size, genre and synopsis.

NBD Biblion is responsible for creating metadata for every book and media title in their collections, maintaining the data in MIPS. While historically, this has been labour intensive, NBD Biblion believes that they can partially automate metadata creation by applying AI techniques. Some years ago, NBD Biblion embarked on a project with Bookarang to automatically extract metadata from book covers and contents.

BSL has worked with Bookarang to share metadata with MIPS, automatically importing metadata to the MIPS databases and returning corrections to improve the AI processes.

Automated metadata using AI

Metadata is a term that may not be familiar to everyone, but metadata refers to characteristics that can describe any object – including books. Metadata can encompass basic details such as the title and author. However, it can also include a wealth of other information, such as the publisher, date of publication, ISBN/EAN number, number of pages, size, genre and synopsis.

NBD Biblion is responsible for creating metadata for every book and media title in their collections, maintaining the data in MIPS. While historically, this has been labour intensive, NBD Biblion believes that they can partially automate metadata creation by applying AI techniques. Some years ago, NBD Biblion embarked on a project with Bookarang to automatically extract metadata from book covers and contents.

BSL has worked with Bookarang to share metadata with MIPS, automatically importing metadata to the MIPS databases and returning corrections to improve the AI processes.

Machine Learning

Sales forecasting using Machine Learning

Building on their earlier project with Enjins, NBD Biblion looked further into the possibilities offered by self-learning systems. They plan to provide Enjins with historic data and metadata from MIPS and apply Machine Learning techniques to determine sales estimates for each title.

We have chosen to develop a new web service for two-way communication between MIPS and Enjins. We share MIPS metadata with Enjins and import their sales forecasts into the MIPS CMS. MIPS users can view the sales estimates, together with the underlying factors that contribute to the forecasts. Access to these factors helps them to better understand the recommendations. So if end-users don’t agree with the automatically determined forecast, they can adjust the figure and provide reasons for their change. Then, we return this information to Enjins so that the Machine Learning model improves. This iterative learning process is key to the principle of Machine Learning. As a result, we hope the system will become increasingly ‘smart’ in determining the results.

Machine Learning

At BSL, you’re at the right address for…

Long-term partnership

BSL has been able to quickly help this client, as we have developed many of their in-house systems and are familiar with their infrastructure and service partners. As new integration possibilities arise, our system architect works with the client and their partners (service providers and third-party developers), designing a suitable integration architecture.

Development expertise

We have developers familiar with many different areas of expertise. These include Web-services (and configuration), database design (SQL Server, Oracle, MySQL), and API developments in Java. We also have Front-end developers with many years of experience in front-end development languages, such as React, JavaScript and PHP.

Continuous improvement

We have made many incremental improvements since we first delivered the Media Information and Production system. These changes usually arise from requests from the NBD Media production team, either adding new features, redesigning interfaces, or re-purposing MIPS content for integration with new products and services or use by third-party developers. We designed MIPS so that it is easy to extend to meet new demands.

“Automating the production of metadata and sales forecasts is an exciting new path on which NBD Biblion has embarked. Thanks to the cooperation with Enjins and their Machine Learning technology, most sales forecasts will be produced automatically. Furthermore, by combining my knowledge of NBD Biblion’s Media Information Production System (MIPS), Enjins’ technology, and NBD Biblion’s vision, BSL will shortly be able to integrate the sales estimates into MIPS.”

Eric

Software Architect, Bright Side of Life

Any questions?

Do you want to get an estimate for developing your custom software. Or advice on creating new websites, or integrating web applications with existing systems. We're happy to answer any questions you may have, no strings attached. Give us a call, or get in touch at sales@brightside.nl!

Share This

    Get in touch

    Contact BSL
    Contact BSL