Editorial assistant
Artificial intelligence (AI) as a smart editorial assistant
Creating glossary entries, setting links in the text or explaining abbreviations - tasks that take up a lot of time for editors. In the case of public institutions, the barrier-free preparation (BITV) of content is also a must, which means additional work.
Based on practical use cases, 3pc has therefore developed an AI-supported editing tool that simplifies editorial workflows: the software application supports editors in linking content, creating accessible content and improving the search function and search results.
Functionalities
- Semantic text analysis based on the latest language technologies (natural language processing)
- Automated recognition and linking of glossary entries and abbreviations (automated indexing)
- Automated recognition and classification of entities such as people, places, organisations and events for semantic search functionalities (semantic search)
- Recommendation system for expanding subject-specific indices (glossary / list of abbreviations)
- Lightweight editor for glossary maintenance (Taxonomy / Ontology)
Benefits
More accessibility
By helping the algorithm to increase links and labelling on the website, accessibility also increases.
Less effort, more time
The machine takes over time-consuming manual tasks - the time saved can be invested elsewhere.
Better ranking
A well-maintained and structured glossary not only enriches your website with background knowledge, but also helps to increase the Google ranking (search engine optimisation / SEO).
Intelligent search
Thanks to the automatic classification of entities, extended search functionalities can be offered, e.g. a faceted search by topic and subject area with a suggestion function or a person search in a semantic relationship network. This increases the quality and accuracy of the results for search queries.
How does the editorial assistant work?
The editing assistant can be integrated into your existing content management solution. It can be easily operated from within the system via a button next to the page content. The wizard analyses the existing content and makes markup suggestions that only need to be checked, adjusted if necessary and confirmed by you.
The editing assistant is systematically organised. The red colour highlighting immediately indicates whether the text has already been marked up. It is also very helpful that abbreviations, entities and the glossary are clearly displayed on different tabs. With one click, you can quickly mark up everything, but also select individual words. The drop-down selection is also easy to use if there are several labelling options.
Use case: the BStU editorial assistant
3pc has implemented an AI-supported tool for editorial work for the Federal Commissioner for Stasi Records (BStU ): an extension for the TYPO3 content management system enables the BStU editorial team to automatically analyse scientific texts and link them to internal knowledge sources in the editorial system. Specifically, AI was used to recognise abbreviations, glossary terms and entities (persons, places, organisations, events).
The special feature: Editors become "curators" by professionally evaluating the machine-generated results - the algorithm learns accordingly and continuously improves.