
How To Detect Toxic Messages in Swedish Language (NLP) - Ebbot Blog
Detecting toxic messages in Swedish language
Even though the rapid development of Internet and social media contributes significantly to human connection, it is undeniable that this is also the very reason why toxic behaviors become more common online. Thus, toxic comments classification has been researched by experts in the Machine Learning field for the past few years. Recently, one of our clients asked us to teach Ebbot to detect toxic messages in conversations. Thanks to this special request, we got a chance to work on one of the most difficult topics in the Natural Language Processing (NLP) field. And yes, we can not be more excited! 🥳
Challenges with collecting dataset
In order to successfully implement this classification task, we have to train Ebbot on a dataset of text with toxicity. Although large labeled training datasets exist, they are not available in Swedish. And using machine translation is not a good approach, since there are many slangs that cannot be translated accurately by machines.
Ebbot's solution to toxic messages detection
After researching, we found an open-source yet highly accurate trained model, built by Laura Hanu at Unitary. In addition to the original version, which only supported English and was trained on Wikipedia comments, Unitary also provided a multilingual model which was trained on 7 different languages (english, french, spanish, italian, portuguese, turkish and russian).
At the same time, we also found a machine translation model by the Language Technology Research Group at the University of Helsinki. This combination enables us to work around the lack of dataset and meet our clients' request. After receiving input text in Swedish, Ebbot will translate it to English first, then run it through the toxicity classifier. The output will be the scores for six categories of toxic messages: toxicity, severe toxicity, obscene, threat, insult and identity hate. Using this method, not only can we decide whether a message is toxic or not, but we are also able to see which type of inappropriate behaviors it brings.
We are aware that this is not the best solution when it comes to solving Machine Learning/Artificial Intelligence problems. Nevertheless, when facing the challenges of not having available training dataset, we consider this to be one of the quickest and easiest ways to tackle multilingual NLP challenges. Currently we are testing the model and gathering user feedback to improve the app's performance. But please feel free to contact us if you have any inquiries about our bot-builder product or special NLP integrations 🙌 We are usually very responsive 😉
More stories

How the EU AI Act will shape the future of service automation
The clock is ticking. The EU AI Act is set to become law, reshaping how artificial intelligence is developed, deployed, and regulated in Europe. For organizations looking to integrate AI solutions, this legislation raises important questions about compliance, accountability, and the choice of AI providers.

Ebbot Achieves ISO 27001 Certification
In 2024, we took on a bold challenge: to earn the internationally recognized ISO 27001 certification. In December, we achieved that goal, marking an important milestone in Ebbot’s commitment to delivering AI-powered service automation with the highest standards of security.

Press release: Gofido first to launch EbbotGPT to customers - Ebbot Blog
Swedish insurance provider Gofido is taking a significant step in its commitment to delivering exceptional customer service by officially launching EbbotGPT. This marks a historic milestone as Gofido becomes the first insurance provider in Sweden to integrate generative AI into its customer support chatbot.

We’re opening our API for EbbotGPT
In celebration of the one-year anniversary of EbbotGPT, we are happy to announce that we are now opening our API for our EU-hosted LLMs, EbbotGPT. This marks a significant milestone in our journey to offer robust AI-driven customer service solutions that are fully compliant with EU data regulations.

From overwhelmed to empowered: GenAI’s role in succeeding with self-service in ITSM
In today’s fast-paced business world, having an efficient internal service management (ITSM) system is more important than ever. But let’s be honest—many ITSM systems are neither user-friendly nor scalable, which ends up making them inefficient. Enter Generative AI (GenAI), a technology that could solve this. But how can we take advantage of this technology in an effective use case without risking security? Let’s break it down.

Ebbot becomes the preferred GenAI partner to renowned chatbot expert Campfire AI
Stockholm, Sweden – July 8, 2024 Campfire AI, a Brussels-based conversational AI consultancy firm, has handpicked Ebbot as its new GenAI partner. From now on, Campfire AI will offer Ebbot’s services to all clients seeking to leverage GenAI in service automation. Ebbot,…

Enento Group chooses Ebbot as strategic AI partner for service automation
Stockholm, Sweden – June 19, 2024 **With a focus on providing a secure GenAI platform for automating service processes at scale, Ebbot has become an attractive partner for enterprises looking to deliver a world class AI service experience. Now signing the Nordic knowledge company [Enento…

Small vs. Large GenAI models – pros & cons
When it comes to generative AI (GenAI) models, size does matter—just maybe not how you'd expect. Both small and large GenAI models have their strengths and weaknesses. Understanding these can help you choose the best model for your needs. Let's break down the pros and cons.🌟 ## The buzz…

Coeo leverages Generative AI to enhance customer experience
coeo Inkassos is rapidly growing and aims to be one of Sweden's largest debt collection agencies in the next five years. Focusing on customer experience as a central strategy, coeo has now set itself apart by becoming the first in the industry to offer 24/7 support with generative AI.

How to make your data sources AI-ready: Step-by-step
Generative AI has revolutionized chatbot training. What once took hours is now completed in minutes. BUT, (there's always a but), the effectiveness of a Generative AI-trained chatbot heavily depends on the quality of its data sources. So, what constitutes a "good" data source for a GenAI chatbot, and what measures can be taken to prepare? Let's find out.

Cross-border service: coeo's live chat breaks down language barriers with a click
The debt collection company coeo Sweden takes its customer service to the next level by introducing an automatic translation feature in its live chat. With the new feature, users can now get real-time support in any language they prefer.

Ebbot Acknowledged by Deloitte as One of the Top 50 Fastest-Growing Technology Companies in Sweden
Stockholm, Sweden, November 2, 2023. Ebbot, providing a conversational AI platform for managing service processes at scale, has been acknowledged by Deloitte as one of the top 50 fastest-growing technology companies in Sweden. ### Background Ebbot,…