Segmentation
Cluster your readers by topics they are interested in, and track and analyze the popularity, not just of individual articles, but of topics.
Artificial intelligence does the tedious work, so journalists can focus on what’s really important.
Tagging media production with keywords is necessary for proper user segmentation, advertisement targeting, good SEO, and much more. But tagging is boring, tedious, and error-prone. Let Geneea Interpretor do it for you automatically.
Cluster your readers by topics they are interested in, and track and analyze the popularity, not just of individual articles, but of topics.
Advertisers can specify which keywords their ads should or should not appear with.
Find related articles easily. Allow readers to follow certain keywords. Automatically find appropriate photos.
Keywords will improve the interconnectedness of your website. New pages will be created for individual keywords.
“Geneea’s keywords have saved us a lot of manual work. We were really impressed with their quality. Moreover, Geneea is going out of their way to fine tune their API to our needs.”
Ask two journalists to assign keywords to the same articles, and on average, they will agree on less than 20% of those keywords.
Our system takes into account your whole media archive, does not get tired, and follows editorial preferences to the letter.
Our system relies on a huge knowledge base containing millions of records. Scientists, athletes, singers, politicians, cell phones, villages, you name it. And of course, we can integrate your existing keywords, use your IDs, etc.
Do you need more or fewer keywords? General or specific ones? Are some keywords absolutely necessary or absolutely forbidden in some context? We have you covered. The system is fully configurable, either globally or by topic.
Our system is constantly learning from your feedback.
When you decline a keyword we suggest or add a new one, use our feedback API to let us know.
We can add IAB or IPTC categories. You can also use our semantic vectors to cluster similar articles together.
Our system is easy to integrate. Just call our industry standard REST API. It is 99.99% available.
Let us know what you need. Our experts will analyze your requirements and suggest the best setup. There is no commitment from your side.
We fine-tune our system to match your needs and to return exactly the keywords you want. We can usually do this in one day or less.
Your custom models are live and ready to be used via our REST API. Call it, use it.
Ideally, we assign keywords to all your old articles as well. It’s good to be consistent. This is done only after you are happy with our results. We can use a dump from your CMS or, with your permission, we can crawl the articles on your website.
This year’s Data Science Day, an international conference organized by WAN-IFRA, was hosted by the prestigious German newspaper Süddeutsche Zeitung in Munich. The event brought together data professionals from digital news, including analysts, scientists, engineers, and chief data officers, to share insights and experiences. One of the thematic roundtables focused on Retrieval-Augmented Generation (RAG) systems,...
The Reuters Institute at Oxford University has released its annual predictions on the future of journalism, particularly in the digital world, for the next 12 months. What key trends in media and AI are highlighted by the research? The report, Journalism, Media, and Technology Trends and Predictions 2025, published annually since 2012, surveyed the opinions...
This week, our co-founder Petr Hamerník attended WAN IFRA’s AI Forum in Paris. This conference connects the biggest innovators in AI and journalism, and we are honored to be part of this inspiring community. So, what are the key insights for implementing AI into journalistic processes in 2024? 1) Generative AI is here to stay—it...