PoolParty Customer Success Stories
Taxonomies, Ontologies & Knowledge Graphs in Practice
Need inspiration for your use case or a better understanding of semantic web technologies? We’ve put together the latest PoolParty customer success stories to help you out.
Sometimes adopting innovative technologies like ours can be a bit overwhelming. But we’ve got your back. This collection of success stories will walk you through different challenges you may relate to and how PoolParty helped solve them.
Digital Marketing Manager
POOLPARTY CUSTOMER SUCCESS STORY #1: MICROSOFT DOCS
Improving the Search Experience in Technical Documentation with Knowledge Graphs
Microsoft has been our customer since 2019. They came to us because they wanted to create a cohesive information architecture for their online documentation (Microsoft Docs) to improve the search & learning experience of every person and every organization working with Microsoft products.
As Dana Bublitz, Senior Information Architect, points out in her presentation at the PoolParty Summit 2021, Microsoft Docs began to grow in complexity, and as this happened, they identified that a lack of structure was affecting what their teams could actually do:
- They couldn’t link content at scale other than manually hard-coding the links.
- They spent significant engineering time and costs manually tagging content.
- They didn’t have an overview of duplicate or outdated content and,
- Their ability to support the user was limited in various scenarios.
At the time, they managed taxonomies in spreadsheets. They realized that one of the main disadvantages of using taxonomies in the form of spreadsheets is the difficulty of obtaining cohesive information, especially when the information is spread among multiple teams, documents and data silos.
They needed a taxonomy management tool to manage a centralized content model for the site, greatly reducing efforts to manage taxonomies and tag content.
With the PoolParty Semantic Suite, they began to classify millions of documents, tag them and make them discoverable, all based on a knowledge graph. The starting point was taxonomies, but soon after they began to enrich them with ontologies to identify interconnections and relationships in their common vocabulary.
Maintaining the knowledge graph in PoolParty has proven to be very efficient, says Dana Bublitz:
Initially, it took their engineering team up to two weeks to edit basic terms or single tags. Because of PoolParty, however, this time window has shrunk to less than 24 hours!
Microsoft Docs uses the PoolParty Semantic Suite on a daily basis and for a variety of purposes. First, to improve not only documentation but also search, contextual navigation, and recommendations. This allows them to offer a better user experience and seamless communication and workflows from different teams and departments.
To help you visualize what the search experience now looks like for the user, take a look at the screenshots below.
In this image from Microsoft Docs, you can see a faceted search that offers multiple filters to narrow down the results based on different criteria. This is extremely useful for helping visitors discover content, especially on content-heavy sites. This makes it easier for Microsoft Docs users to find “the needle in the haystack”.
In both cases, the information you see as autocomplete suggestions or the criteria in the faceted search comes from the taxonomies created in PoolParty. That means information can be traced back to the controlled vocabulary created in PoolParty and the way it was organized using taxonomies, ontologies, and knowledge graphs. Isn’t that amazing?
From Unknowable Chaos to Knowledge Graph
Dana Bublitz, Senior Information Architect at Microsoft
Watch Dana’s presentation at the PoolParty Summit 2021 where she explains the whole journey Microsoft Docs went through to get from unknowable chaos to a knowledge graph thanks to PoolParty Semantic Suite. If you want to have an even better overview of this use case, don’t forget to download the customer success story.
POOLPARTY CUSTOMER SUCCESS STORY #2: JOBTEASER
Connecting Data with Ontologies for a Tinder-like Job Swipe Platform
JobTeaser, a French recruitment platform that is rapidly expanding to other European countries, has been our client since 2020. They came to us because they needed to easily integrate and distribute internal and external data to deliver value to their users with personalized recommendations. Furthermore, considering their wide range of services and applications, they needed a tool to centrally model their business domain with a semantically enriched controlled vocabulary in multiple languages.
From the perspective of the Senior Information Architect:
There would be no way to integrate the information into our products [without PoolParty] because all the data points we measure for students are outside the [JobTeaser] ecosystem.
In PoolParty, they use concepts to build the controlled vocabulary. Concepts are a sophisticated alternative to simple terms because they label and organize objects with a variety of semantic components, such as synonyms or the preferred name for the concept. You can also add images, georeferences, descriptions, hierarchies, and relationships to a concept in PoolParty.
As a result, JobTeaser considers the controlled vocabulary in PoolParty as the “orchestrator and source of truth” that JobTeaser applications use to calculate a user’s scores to facilitate the matching process. Take a look at the downloadable story to understand how it works.
In addition, PoolParty and its linked data capabilities make it easy to improve scoring metrics and deliver more accurate matches by combining JobTeaser’s own taxonomy with ESCO, the multilingual official taxonomy of common skills, qualifications and occupations in Europe.
Along with concepts in the taxonomies, JobTeaser uses powerful ontologies in PoolParty. Ontologies help them add a layer of semantic expressiveness to their data and trace complex relationships. This makes it easier, for example, to model relations between professional domains, hard skills and soft skills to determine if a person will be a good fit for a specific role. As a result, the recommended jobs are truly relevant and useful to students using the Tinder-like job swipe platform.
In the above screenshots taken from François’s presentation at the PoolParty Summit 2021, you can see Marco, one of JobTeaser’s psychometric tools on the left, and the taxonomy created in PoolParty on the right. Marco is a questionnaire that assesses the students’ professional interests based on different dimensions, such as organizer, innovator, pragmatist, etc. to deliver a user profile. Each question is referenced to the dimensions, also known as concepts, stored in PoolParty. These concepts have been enhanced with properties in the ontology to express the semantic nuances of a profile. In this way, JobTeaser can offer personalized recommendations from the most to the least outstanding dimensions that a student has obtained in the resulting profile after completing the questionnaire.
An Ontology for a Universal Guidance Journey
François Violette, Senior Information Architect at JobTeaser
Watch François’s presentation at the PoolParty Summit 2021 where he explains how PoolParty Semantic Suite is the underlying technology to connect their data with semantics. This allows JobTeaser to offer personalized recommendations to students and young graduates that matches their user profiles. If you want to have an even better overview of this use case, don’t forget to download the customer success story.
POOLPARTY CUSTOMER SUCCESS STORY #3: WOLTERS KLUWER
Taxonomies & Metadata as the Backbone for Digital Transformation
Wolters Kluwer has been our customer for over a decade and we have been helping them on their digital transformation journey ever since.
They came to us with the goal of diversifying their product offerings to serve legal professionals in law firms, general counsel offices, and corporate legal departments with data-driven decision-making tools.
The tools that Wolters Kluwer offers its customers are largely based on the creation, processing and dissemination of content. As you can imagine, they have accumulated vast amounts of content over the years that was stored in XML. While XML is good at processing data, it fails at enriching it with metadata. They were looking for a solution to systematically separate metadata from the text itself and manage controlled vocabularies to facilitate dynamic content publishing, including linked data capabilities to connect internal and external content sources.
As a technology based on the W3C standards, PoolParty was a good match for Wolter Kluwer’s digital transformation process. Poolparty’s user-friendly interface made it easy for legal experts to transfer their knowledge in a technical environment. This was an important factor in building and maintaining powerful domain taxonomies that are now a strong backbone for other tasks such as text mining.
This new semantic infrastructure helped Wolters Kluwer increase collaboration with other teams and geographic locations within the same company.
The digital transformation process that the company started more than 10 years ago has caused a cultural change in Wolters Kluwer. Now they are no longer focused on content, but their whole data ecosystem, as the new implemented system is scalable, promotes collaboration between teams and helps connect content with valuable data, such as customer information, which opens up a new range of possibilities.
Wolter Kluwers’ online tools help to streamline legal and regulatory research, analysis, and workflows to drive value to organizations, ensuring more transparent, just and safe societies (as stated on their website). In the screenshot above, you can see how their German division offers an online library of resources including faceted search and search autocompletion powered by taxonomies created and maintained in the PoolParty Semantic Suite.
Christian Dirschl , Chief Content Architect at Wolters Kluwer
Watch Christian’s testimonial interview to experience Wolters Kluwer’s journey towards a semantically powered digital transformation. If you want to have an even better overview of this use case, don’t forget to download the customer success story.